TW201944338A - Data processing method, apparatus and device for insurance fraud identification, and server - Google Patents

Data processing method, apparatus and device for insurance fraud identification, and server Download PDF

Info

Publication number
TW201944338A
TW201944338A TW108104900A TW108104900A TW201944338A TW 201944338 A TW201944338 A TW 201944338A TW 108104900 A TW108104900 A TW 108104900A TW 108104900 A TW108104900 A TW 108104900A TW 201944338 A TW201944338 A TW 201944338A
Authority
TW
Taiwan
Prior art keywords
data
relationship
identified
insurance
personnel
Prior art date
Application number
TW108104900A
Other languages
Chinese (zh)
Other versions
TWI686760B (en
Inventor
王修坤
鄒曉川
Original Assignee
香港商阿里巴巴集團服務有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 香港商阿里巴巴集團服務有限公司 filed Critical 香港商阿里巴巴集團服務有限公司
Publication of TW201944338A publication Critical patent/TW201944338A/en
Application granted granted Critical
Publication of TWI686760B publication Critical patent/TWI686760B/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/28Databases characterised by their database models, e.g. relational or object models
    • G06F16/284Relational databases
    • G06F16/288Entity relationship models
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/901Indexing; Data structures therefor; Storage structures
    • G06F16/9024Graphs; Linked lists
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/08Insurance
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2216/00Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
    • G06F2216/03Data mining

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • General Physics & Mathematics (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Engineering & Computer Science (AREA)
  • Technology Law (AREA)
  • General Business, Economics & Management (AREA)
  • Software Systems (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

A data processing method, apparatus and device for insurance fraud identification, and a server, wherein multi-scale relationship network graph data of a crowd are built on the basis of multi-scale relationship association data of insurance applicants and insurants, a relationship network among people may be more deeply mined, identification efficiency is improved, and identification range is widened. At the same time, a supervised learning model is jointly built according to characteristic data of a fraudster and is used for learning relationship network characteristics and personal characteristics of the fraudster. Accomplice fraudsters will have obvious multi-scale relationship characteristics in the relationship network, the characteristics of the fraudsters frequently indicate similarities, thus the fraudsters may be more effectively and efficiently identified by using the described method, and identification processing efficiency is improved.

Description

保險欺詐識別的資料處理方法、裝置、設備及伺服器Data processing method, device, equipment and server for identifying insurance fraud

本說明書實施例方案屬保險反欺詐識別的電腦資料處理的技術領域,尤其涉及一種保險欺詐的資料處理方法、裝置、處理設備及伺服器。The solutions in the embodiments of the present specification belong to the technical field of computer data processing for insurance anti-fraud identification, and more particularly, to a data processing method, device, processing equipment, and server for insurance fraud.

保險是透過繳納規定的保費,然後可以享受的財務、人身等保障。隨著社會的經濟發展和人們保險意識的提高,保險業務的需求也越來越多。
然而,由於保險有一定的經濟槓桿效應,使得市場上出現大量騙保的行為,這些騙保人員通常故意製造保險事並依此獲得保險公司賠款。目前的騙保行為有發展為專業化、團隊化的趨勢,對保險行業的健康發展帶來非常不利的影響,損壞保險公司和公眾利益。目前傳統的識別騙保的方式主要依靠任人工利用一些簡單規則對歷史騙保人員進行識別,憑藉歷史騙保人員的行為預測是否存在騙保風險。由於騙保人員和團體的隱蔽性越來越強,現有的這種方式不容易快速發現團體作案,並且人工審核的工作量較大,識別效率較為低下。
因此,業內極需一種可以更加有效和高效的識別出騙保人員的處理方式。
Insurance is through the payment of prescribed premiums, and then you can enjoy financial, personal and other protections. With the economic development of the society and the improvement of people's insurance awareness, the demand for insurance business is also increasing.
However, due to the certain economic leverage effect of insurance, a large number of fraudulent acts have occurred in the market. These fraudulent personnel usually deliberately create insurance matters and obtain insurance company compensation accordingly. The current insurance scams have developed into a professional and team-oriented trend, which has a very adverse impact on the healthy development of the insurance industry and damages the interests of insurance companies and the public. At present, the traditional way to identify fraudulent insurance is to rely on any human to use some simple rules to identify historical fraudulent insurance personnel, and rely on the behavior of historical fraudulent insurance personnel to predict whether there is a fraudulent insurance risk. Due to the increasing concealment of insurance scammers and groups, it is not easy to quickly detect group crimes in this way, and the workload of manual review is large, and the identification efficiency is relatively low.
Therefore, there is a great need in the industry for a processing method that can more effectively and efficiently identify fraudsters.

本說明書實施例目的在於提供一種保險欺詐的資料處理方法、裝置、處理設備及伺服器,可以提供利用人員之間的關係網路資料和自身特徵,可以更加有效的識別出騙保人員。
本說明書實施例提供的一種保險欺詐的資料處理方法、裝置、處理設備及伺服器是包括以下方式實現的:
獲取待識別人群的關係關聯資料;
基於所述關係關聯資料構建所述待識別人群的多度關係網路圖資料以及提取所述待識別人群的人員特徵資料;
利用構建的有監督學習算法對所述待識別人群的多度關係網路圖資料和所述人員特徵資料進行識別,確所述待識別人群騙保輸出結果;所述有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。
一種保險欺詐識別的資料處理裝置,包括:
資料獲取模組,用於獲取待識別人群的關係關聯資料;
特徵計算模組,用於基於所述關係關聯資料構建所述待識別人群的多度關係網路圖資料以及提取所述待識別人群的人員特徵資料;
欺詐識別模組,用於利用構建的有監督學習算法對所述待識別人群的多度關係網路圖資料和所述人員特徵資料進行識別,確所述待識別人群騙保輸出結果;所述有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。
一種處理設備,包括處理器以及用於儲存處理器可執行指令的記憶體,所述處理器執行所述指令時實現:
獲取待識別人群的關係關聯資料;
基於所述關係關聯資料構建所述待識別人群的多度關係網路圖資料以及提取所述待識別人群的人員特徵資料;
利用構建的有監督學習算法對所述待識別人群的多度關係網路圖資料和所述人員特徵資料進行識別,確所述待識別人群騙保輸出結果;所述有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。
一種伺服器,包括至少一個處理器以及用於儲存處理器可執行指令的記憶體,所述處理器執行所述指令時實現:
獲取待識別人群的關係關聯資料;
基於所述關係關聯資料構建所述待識別人群的多度關係網路圖資料以及提取所述待識別人群的人員特徵資料;
利用構建的有監督學習算法對所述待識別人群的多度關係網路圖資料和所述人員特徵資料進行識別,確所述待識別人群騙保輸出結果;所述有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。
本說明書實施例提供的一種保險欺詐的資料處理方法、裝置、處理設備及伺服器,基於投保人員和被保險人的多維度的關係關聯資料構建人群的多度關係網路圖資料,可以更加深入的挖掘人員之間的關係網路,提高識別效率和範圍。同時結合騙保人員自身的特徵資料,共同建立有監督的學習模型,用來學習騙保人員的關係網路特徵和自身特徵。團夥的騙保人員不僅在關係網路上有著較為明顯和多度的關係特徵,其自身特徵也常常表現出相似性,因此利用本說明書實施例提供的方法可以更加有效和高效的識別出騙保人員,提高識別處理效率。
The purpose of the embodiments of the present specification is to provide a data processing method, device, processing device and server for insurance fraud, which can provide network data and self-characteristics of the relationship between personnel, and can more effectively identify fraud insurance personnel.
The method, device, processing equipment and server for insurance fraud data provided by the embodiments of this specification are implemented in the following ways:
Obtain relationship data of the population to be identified;
Constructing a multi-degree relationship network diagram data of the crowd to be identified based on the relationship association data and extracting person characteristic data of the crowd to be identified;
The constructed supervised learning algorithm is used to identify the multiple degree network graph data of the crowd to be identified and the characteristics of the person, and to confirm the fraudulent output of the crowd; the supervised learning algorithm includes using The data relationship model of the selected target population based on the multi-relationship network data and personnel characteristics data, and the marked historical fraud insurance personnel are trained as sample data.
A data processing device for identifying insurance fraud includes:
A data acquisition module for acquiring relationship data of a population to be identified;
A feature calculation module, configured to construct abundance relationship network graph data of the population to be identified based on the relationship association data and extract person characteristic data of the population to be identified;
A fraud identification module is configured to identify the abundance relationship network graph data of the crowd to be identified and the personnel characteristic data by using a constructed supervised learning algorithm, and confirm the fraudulent output of the crowd to be identified; Supervised learning algorithms include data relationship models that are trained using sampled target group multiple-relationship network data and personnel characteristics data, and marked historical fraud insurance personnel as sample data.
A processing device includes a processor and a memory for storing processor-executable instructions. When the processor executes the instructions, the processor implements:
Obtain relationship data of the population to be identified;
Constructing a multi-degree relationship network diagram data of the crowd to be identified based on the relationship association data and extracting person characteristic data of the crowd to be identified;
The constructed supervised learning algorithm is used to identify the multiple degree network graph data of the crowd to be identified and the characteristics of the person, and to confirm the fraudulent output of the crowd; the supervised learning algorithm includes using The data relationship model of the selected target population based on the multi-relationship network data and personnel characteristics data, and the marked historical fraud insurance personnel are trained as sample data.
A server includes at least one processor and a memory for storing processor-executable instructions. When the processor executes the instructions, the server implements:
Obtain relationship data of the population to be identified;
Constructing a multi-degree relationship network diagram data of the crowd to be identified based on the relationship association data and extracting person characteristic data of the crowd to be identified;
The constructed supervised learning algorithm is used to identify the multiple degree network graph data of the crowd to be identified and the characteristics of the person, and to confirm the fraudulent output of the crowd; the supervised learning algorithm includes using The data relationship model of the selected target population based on the multi-relationship network data and personnel characteristics data, and the marked historical fraud insurance personnel are trained as sample data.
A method, device, processing device and server for insurance fraud data provided in the embodiments of the present specification. Based on multi-dimensional relationship data of the insured and the insured, a multi-degree relationship network diagram of the crowd can be further developed. A network of relationships among miners to improve identification efficiency and scope. At the same time, combined with the characteristics of the fraudsters themselves, a supervised learning model is jointly established to learn the characteristics of the relationship network and the characteristics of the fraudsters. Gang fraudsters not only have obvious and abundant relationship characteristics on the relationship network, but their own characteristics often also show similarities. Therefore, the methods provided in the examples of this specification can more effectively and efficiently identify fraudsters To improve the efficiency of recognition processing.

為了使本技術領域的人員更好地理解本說明書中的技術方案,下面將結合本說明書實施例中的附圖,對本說明書實施例中的技術方案進行清楚、完整地描述,顯然,所描述的實施例僅僅是本說明書中的一部分實施例,而不是全部的實施例。基於本說明書中的一個或多個實施例,本領域普通技術人員在沒有作出創造性勞動前提下所獲得的所有其他實施例,都應當屬於本說明書實施例保護的範圍。
物以類聚,人以群分。騙保人群通常需要多人配合才能提高騙保的偽裝性。而騙保人員的聚集在很多情況下也會基於熟人關係或具有較為明顯的共性特徵或某一維度的網路關係特徵資料。例如以親戚之間合夥的騙保行為,傳銷性質的具有明顯階層劃分的騙保團體、有經驗的歷史騙保人員為頭目拉攏的社會群體或學生群體等。本說明書實施例提供多個實施方案中,從包含投保人員和申請理賠人員的目標人群的多種關係關聯資料觸發,進行多度關係網路的構圖(關係網路圖的資料可以稱為多度關係圖資料),深入挖掘目標人群之間的關係網路,解決有常規僅對歷史騙保人員和與歷史騙保人員有直接關係的一度關係進行識別的覆蓋率和識別率低的問題。同時,本說明書實施例提供的方案,還考慮到騙保人員自身的特徵屬性,如騙保人員通常使用虛假資訊註冊帳號、帳號註冊時間短、帳號註冊後主使用投保業務等。本說明書提供的實施方案,結合騙保團體的關係特徵資料和自身特徵資料,將歷史騙保人員標記出來,進行有監督模型的算法學習,從而可以計算或識別出待識別人群是否存在騙保的結果。
下面以一個具體的保險業務欺詐識別處理的應用場景為例對本說明書實施方案進行說明。具體的,圖1是本說明書提供的所述一種保險欺詐識別的資料處理方法實施例的流程示意圖。雖然本說明書提供了如下述實施例或附圖所示的方法操作步驟或裝置結構,但基於常規或者無需創造性的勞動在所述方法或裝置中可以包括更多或者部分合併後更少的操作步驟或模組單元。在邏輯性上不存在必要因果關係的步驟或結構中,這些步驟的執行順序或裝置的模組結構不限於本說明書實施例或附圖所示的執行順序或模組結構。所述的方法或模組結構的在實際中的裝置、伺服器或終端產品應用時,可以按照實施例或者附圖所示的方法或模組結構進行順序執行或者並行執行(例如並行處理器或者多執行緒處理的環境、甚至包括分布式處理、伺服器集群的實施環境)。
當然,下述實施例的描述並不對基於本說明書的其他可擴展到的技術方案構成限制。例如其他的實施場景中,本說明書提供的實施方案同樣可以應用到基金欺詐識別、產品交易、服務交易等的實施場景中。具體的一種實施例如圖1所示,本說明書提供的一種保險欺詐識別的資料處理方法可以包括:
S0:獲取待識別人群的關係關聯資料;
S2:基於所述關係關聯資料構建所述待識別人群的多度關係網路圖資料以及提取所述待識別人群的人員特徵資料;
S4:利用構建的有監督學習算法對所述待識別人群的多度關係網路圖資料和所述人員特徵資料進行識別,確所述待識別人群騙保輸出結果;所述有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型
本實施例應用場景中,通常保險出險、核算、賠付等環節主要針對的是申請理賠人員,本說明書實施例中考慮了一些實際場景中騙保動機存在從投保開始就產生的情況,騙保人員主要目的是為了申請獲得保險賠付金額,當然也有一些在投保之後才有的騙保動機。被保險人為出險的主要主體,如老鄉團體的騙保人員故意製造被保險人人的意外事故因此本實施例的在識別是否存在騙保時的目標人群時選取了申請理賠人員和被保險人的人員集合。因此,本說明書所述方法的一些實施例中,選取目標人群進行關係特徵資料的獲取學習時,所述的目標人群可以包括申請理賠人員和被保險人的人員集合。需要說明的是,申請理賠人員在一些實施情況下可以包括投保人,如父親給兒子投保,父親為受益人,出險後父親為申請理賠人員;或者一些實施情況下申請理賠人員也可能包括被保險人員,如投保人給自己的投保,受益人為自己。上述中所述的申請理賠人員和被保險人可以理解的是保險業務中處於不同角色的人員類別名稱,並不現在是不同的人員,一些實施場景中所示的申請理賠人員和被保險人員可以全部或部分相同。
當然,其他的實施例中,目標人群的選取也可以選取理賠申請人員或投保人或被保險人或受益人等中的一種或多種。
所述的關係關聯資料可以包括多種維度的與所述目標人群中人員相關聯的資料資訊,如戶籍、年齡、人員之間的親屬/同學關係、投保資料、保險出險資料等等。具體的關係關聯資料可以根據實際的應用場景中進行選取確定使用哪些類別的哪些資料,一般的,作業人員可以根據騙保行為可能涉及到的資料資訊作為採集關係關聯資料的依據。本說明書提供的一個實施例中,所述的關係關聯資料可以包括下述中的至少一種:
社會關係資料、終端資料、終端的應用以及應用帳戶操作資訊、與保險行為關聯的行為資料、人員基礎屬性資料、地理位置資料。
所述的社會關係資料可以包括目標人群中人員之間的社會關係,如堂兄弟、師生、家人、同學、領導與下屬等。所述的終端資料可以包括人員使用的通信設備的品牌、型號、類別,一些騙保場景中人員使用相同品牌的手機。終端的應用以及應用帳戶操作資訊,可以用於確定是否使用同一款應用,以及使用相同的帳戶登錄不同終端的應用進行保險欺詐操作,一些場景中多個下述聽從頭目統一指揮在終端上應用上進行操作。所述的與保險行為關聯的行為資料可以包括目標人的投保行為、理賠行為、賠償金額等行為資料。所述的人員基礎屬性資料可以包括投保人/申請理賠人員的年齡、性別、職業、戶籍等。所述的地理位置資料可以包括目標人群當前所處的地理位置資訊或者歷史到過/滯留果的區域的資訊。當然,上述所述的各個維度的資料關係關聯資料還可以有其他的定義或包含更多/更少的資料類別和資訊,也可以包括除上述之外的其他維度的關係關聯資料,如消費資訊甚至信用記錄或行政處罰資訊,具體採集時可以採集上述中的一種或多種資料資訊。
所述的人員特徵資料可以包括與單個人員自身相關聯的資料資訊,如性別、年齡、保險服務帳號或終端應用帳戶註冊時間、信用記錄、消費情況等,或者還可以包括與保險行為關聯的行為資料,如多次投保行為、經常性的理賠行為、賠償金額是否正常等。還可以包括以下其他的商品或服務的交易資料,如長期的大額支出,多次出車險,購買多部手機,註冊多個通信帳號/社交帳號等。
具體的人員特徵計算使用的人員特徵資料可以採用上述中的一種或多種的組合,以實現人員自身特徵的識別。因此,另一個實施例中,所述人員特徵資料可以包括用戶註冊帳號、交易資料、與保險行為關聯的行為資料中的至少一項提取出來的特徵資料。
騙保團夥的人員之間通常存在較為緊密的關係網路,本實施例中可以利用上述獲取的多維度的關係關聯資料構建目標人群的多度關係網路圖資料。所述的多度關係網路圖資料可以包括基於所述關係關聯資料建立的不同人員之間的關係鏈而生成的關係網路圖,其中的關係網路圖上人員之間的關係鏈資料為多度關係網路圖資料。所述的關係鏈可以表示每兩個人員之間的關係資料,如A與B是老闆關係、A與C是家人關係等。單獨的兩個人員之間的關係可以稱為一度關係,本實施例中所述的多度關係網路圖資料中的“多度”可以包括基於所述一度關係建立的新的人員之間的關聯資料,如基於第一人員與第二人員的一度關係和第二人員與第三人員的一度關係建立的所述第一人員與第三人員的二度關係,甚至進一步可以基於其他一度關係建立第一人員與第四人員的三度關係等等。
如一個示例中,A是單個人員,B是A的姐夫,則A與B是一度的社會關係,A與其姐夫B的公司老闆C之前不存在社會關係,但在本說明書實施例中,由於存在B既是A的姐夫又是公司老闆C的下屬,因此A與公司老闆C之間建立的二度關係。
除上述人員之間的社會關係之外,還可以根據採用的關係關聯資料或者關係構建需求形成其他類型的多度關係網路圖資料,如是否為老鄉,使用同一種通信工具、多人終端上的某個應用在固定時間段登錄等。當然,基於所述關係關聯資料構建關係網路具體的實現中,關係之間的確定可以預先設計成立關係鏈的規則。
基於構建好的多度關係網路圖資料和提取的人員特徵資料,本實施例可以採用有監督的學習算法學習騙保人員的關係特徵和自身特徵,從而可以建立有效的識別模型。
通常的,機器學習的常用方法主要分為有監督學習,有時也簡稱監督學習(supervised learning)和無監督學習(unsupervised learning)。監督學習是一種分類處理方式,通常針對有標簽的資料集,透過已有的訓練樣本(即已知資料以及其對應的輸出)去訓練得到一個最優模型(這個模型屬於某個函數的集合,最優則表示在某個評價準則下是最佳的),再利用這個模型將所有的輸入映射為相應的輸出,對輸出進行簡單的判斷從而實現分類的目的,也就具有了對未知資料進行分類的能力。監督學習裡典型的例子就是KNN(k-NearestNeighbor,鄰近算法)、SVM(Support Vector Machine),支持向量機)。有監督學習算法在有一定數量的訓練樣本的情況下,相比於無監督算法可以得到更為準確的輸出結果。
根據選取的不同的有監督學習算法,其他具體的關係特徵和自身特徵的處理過程根據算法種類和識別處理需求進行設計和確定。例如可以採用Structure2vec等的有監督圖算法。例如一個實施例中,所述構建的有監督學習算法包括:
S40:利用選取的有監督學習算法對目標人群的多度關係網路資料中目標人員與其他人員的關係特徵進行第一關係網路學習、基於所述目標人員特徵的自身特徵資料進行第二自身屬性學習;
S42:以所述第一關係網學習和第二自身屬性學習得到的特徵資料作為所述有監督學習算法的自變量,以打標的歷史騙保人員作為因變量建立關係模型;
S44:在所述關係模型的輸出達到預設準確率時確定構建的有監督學習算法。
圖2是本說明書提供的一種構建有監督學習算法實施例的處理過程示意圖。
如圖2所述的示例中,可以使用Structure2vec的有監督圖算法:一方面去學習目標人及其鄰居的關係特徵(如與多少人有關係,是否跟騙保人員有關係),另一方面學習目標人本身的特徵(如性別、年齡等),以上特徵作為模型的x變量;其次,根據歷史打標好是否是騙保人員作為y變量;最後,根據y和x建立相關模型,從而達到僅依靠x就可以預測y情況。
本實施例應用場景中所述的最終識別出是否為騙保的可以是單獨的一個人。即本實施例中的理由有監督學習算法學習了團夥騙保的關係特徵之後,再結合騙保人員自身的特徵,可以直接得到某個待識別人員是否為騙保人員或者是騙保人員的概率的騙保輸出結果。如可以為人員打標為騙保人員或正常人員,或者為騙保人員的概率。
當然,這裡所述的標記為騙保人員是基於關係特徵和自身特徵的識別結果,可以作為初步確定這些人是否為騙保人員的依據和參考。最終確定是否為騙保時可以有作業人員主觀判斷,或者再結合其他的計算方式進行判斷和確定。
本實施例提供的保險欺詐的資料處理方法,可以基於投保人員和被保險人的多維度的關係關聯資料構建人群的多度關係網路圖資料,可以更加深入的挖掘人員之間的關係網路,提高識別效率和範圍。同時結合騙保人員自身的特徵資料,共同建立有監督的學習模型,用來學習騙保人員的關係網路特徵和自身特徵。團夥的騙保人員不僅在關係網路上有著較為明顯和多度的關係特徵,其自身特徵也常常表現出相似性,因此利用本說明書實施例提供的方法可以更加有效和高效的識別出騙保人員,提高識別處理效率。
本說明書提供的所述方法的另一個實施例中,還可以利用歷史騙保人員的資料資訊結合多度關係網路圖資料進行騙保人員的識別。具體的,本說明書提供的所述方法的另一個實施例中,所述關係關聯資料還可以包括:歷史騙保人員名單資料。
本實施例中加入歷史騙保人群的資料資訊,在對所述分類社群進行分析處理時,考慮歷史騙保人員的參與程度。一般的,若歷史騙保人員在某個分類社群中的關係濃度較高,則該分類社群中的人員進行騙保的可能性就越大。本實施例中所述的關係濃度可以包括歷史騙保人員的參與程度,具體的可以包括分類社群中歷史騙保人員的數量、歷史騙保人員的數量占比、歷史騙保人與其他人員的關係密程度等。所述的關係密集程度的一個示例如,10個人員的風險社群中,2個歷史騙保人員與其他6個人員是一度或多度關係的親屬關係,與2個人員是同學關係,則表示可能為傳銷性質的騙保團夥。具體的關係濃度可以採用不同的方式計算,如上述歷史騙保人員數量,占比,關係網路等。本說明書實施例提供另一種實施例中,可以從待識別人群的規模和歷史騙保人員的數量兩個指標來計算所述關係濃度,所述的關係濃度可以作為衡量騙保的概率取值。具體的,可以包括:
以所述待識別人群的人員數量取對數後作為第一因子;
以所述待識別人員中歷史騙保人員的數量占比作為第二因子;
基於所述第一因子與所述第二因子的乘積作為待識別人群的團體騙保概率。
然後可以結合自身特徵計算得到的個人騙保概率取值,與團體騙保概率進行運算來確定最終輸出的團體為騙保或單個人員為騙保的概率。或者所述的團體騙保概率和個人騙保概率分別各自利用,不進行相互計算。
例如,具體實現時,可以採用下述方式計算社群騙保的概率:
RiskDegree=log(分類社群人員總數)*歷史騙保人員數量/分類社群人員總數。
當然,還可以採用其他的計算方式或變形、變換的方式,如取自然對數等,在此限制和贅述。
上述實施例提供了可以利用歷史騙保人員的資料資訊來識別騙保的欺詐群體。本說明書提供的另一種實施例中,可以利用人群中各個成員之間的關係網路特徵來確定是否為騙保人員。具體的,如確定人群中人員關係的網路結構特徵;
若所述網路結構特徵符合預設的騙保網路結構,則將所述人群標記為欺詐群體。
所述的上述方式可以用於有監督學習算法的訓練中,所述的人群為目標人員。對於識別待識別人員的處理中,所述的人群為所述待識別人群。
所述的網路結構特徵可以基於人群中的人員資訊、人員之間的關係網路資訊等。這裡的關係網路資訊可以為前述所述的一度資訊,也可以包括構建的多度資訊。
可以使用一定的算法識別分析社群中關係網路是什麼特徵,如果網路結構特徵符合騙保團夥特徵,此時可以標記為欺詐群體。例如一個示例中,人群中的關係網路可以為比如“球形網路”、“金字塔形網路”等網路結構。“金字塔網路”類似於傳銷組織,一層一層關系結構,屬於騙保的可能性較大;“球形網路”就是網路中彼此關聯,可能為非中心化的騙保組織。
本說明書實施例提供的一種保險欺詐的資料處理方法、使用接近實際關係網路的關係關聯資料支撐關係網路算法的挖掘,實現多度關係的關係網路資料計算。基於投保人員和被保險人的多維度的關係關聯資料構建人群的多度關係網路圖資料,可以更加深入的挖掘人員之間的關係網路,提高識別效率和範圍。同時結合騙保人員自身的特徵資料,共同建立有監督的學習模型,用來學習騙保人員的關係網路特徵和自身特徵。團夥的騙保人員不僅在關係網路上有著較為明顯和多度的關係特徵,其自身特徵也常常表現出相似性,因此利用本說明書實施例提供的方法可以更加有效和高效的識別出騙保人員,提高識別處理效率。
上述所述的方法可以用於客戶端一側的保險欺詐識別,如行動終端安裝反欺詐應用、支付應用提供的保險業務。所述的客戶端可以為PC(personal computer)機、伺服器、工控機(工業控制電腦)、行動智慧電話、平板電子設備、便攜式電腦(例如筆記本電腦等)、個人數位助理(PDA)、或桌面型電腦或智慧穿戴設備等。行動通信終端、手持設備、車載設備、可穿戴設備、電視設備、計算設備。也可以應用在保險業務方或服務方或第三方機構的系統伺服器中,所述的系統伺服器可以包括單獨的伺服器、伺服器集群、分布式系統伺服器或者處理設備請求資料的伺服器與其他相關聯資料處理的系統伺服器組合。
本說明書實施例所提供的方法實施例可以在行動終端、電腦終端、伺服器或者類似的運算裝置中執行。以運行在伺服器上為例,圖3是本發明實施例的一種識別車輛受損部件的伺服器的硬體結構框圖。如圖3所示,伺服器10可以包括一個或多個(圖中僅示出一個)處理器102(處理器102可以包括但不限於微處理器MCU或可編程邏輯裝置FPGA等的處理裝置)、用於儲存資料的記憶體104、以及用於通信功能的傳輸模組106。本領域普通技術人員可以理解,圖3所示的結構僅為示意,其並不對上述電子裝置的結構造成限定。例如,伺服器10還可包括比圖3中所示更多或者更少的組件,例如還可以包括其他的處理硬體,如資料庫或多級緩存,或者具有與圖3所示不同的配置。
記憶體104可用于儲存應用軟體的軟體程式以及模組,如本發明實施例中的搜索方法對應的程式指令/模組,處理器102透過運行儲存在記憶體104內的軟體程式以及模組,從而執行各種功能應用以及資料處理,即實現上述導航互動界面內容展示的處理方法。記憶體104可包括高速隨機記憶體,還可包括非揮發性記憶體,如一個或者多個磁性儲存裝置、閃存、或者其他非揮發性固態記憶體。在一些實例中,記憶體104可進一步包括相對於處理器102遠端設置的記憶體,這些遠端記憶體可以透過網路連接至電腦終端10。上述網路的實例包括但不限於網際網路、企業內部網、局域網、行動通信網及其組合。
傳輸模組106用於經由一個網路接收或者發送資料。上述的網路具體實例可包括電腦終端10的通信供應商提供的無線網路。在一個實例中,傳輸模組106包括一個網路適配器(Network Interface Controller,NIC),其可透過基站與其他網路設備相連從而可與網際網路進行通訊。在一個實例中,傳輸模組106可以為射頻(Radio Frequency,RF)模組,其用於透過無線方式與網際網路進行通訊。
基於上述所述的設備型號識別方法,本說明書還提供一種保險欺詐識別的資料處理裝置。所述的裝置可以包括使用了本說明書實施例所述方法的系統(包括分布式系統)、軟體(應用)、模組、組件、伺服器、客戶端等並結合必要的實施硬體的設備裝置。基於同一創新構思,本說明書提供的一種實施例中的處理裝置如下面的實施例所述。由於裝置解決問題的實現方案與方法相似,因此本說明書實施例具體的處理裝置的實施可以參見前述方法的實施,重複之處不再贅述。儘管以下實施例所描述的裝置較佳地以軟體來實現,但是硬體,或者軟體和硬體的組合的實現也是可能並被構想的。具體的,如圖4所示,圖4是本說明書提供的一種保險欺詐識別的資料處理裝置實施例的模組結構示意圖,可以包括:
資料獲取模組101,可以用於獲取待識別人群的關係關聯資料;
特徵計算模組102,可以用於基於所述關係關聯資料構建所述待識別人群的多度關係網路圖資料以及提取所述待識別人群的人員特徵資料;
欺詐識別模組103,可以用於利用構建的有監督學習算法對所述待識別人群的多度關係網路圖資料和所述人員特徵資料進行識別,確所述待識別人群騙保輸出結果;所述有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。
所述裝置的具體的一個實施例中,所述關係關聯資料可以包括下述中的至少一種:
社會關係資料、終端資料、終端的應用以及應用帳戶操作資訊、與保險行為關聯的行為資料、人員基礎屬性資料、地理位置資料。
所述裝置的另一個實施例中,所述欺詐識別模組103確所述待識別人群騙保輸出結果包括輸出單個待識別目標人員是否為欺詐人員或為欺詐人員的概率。
所述裝置的另一個實施例,所述選取的目標人群包括申請理賠人員和被保險人的人員集合。
所述裝置的另一個實施例,所述人員特徵資料包括用戶註冊帳號、交易資料、與保險行為關聯的行為資料中的至少一項提取出來的特徵資料。
圖5是所述裝置的另一個實施例中,如圖5所示,所述欺詐識別模組103包括:
特徵學習模組1031,可以用於利用選取的有監督學習算法對目標人群的多度關係網路資料中目標人員與其他人員的關係特徵進行第一關係網路學習、基於所述目標人員特徵的自身特徵資料進行第二自身屬性學習;
關係建立模組1032,可以用於以所述第一關係網學習和第二自身屬性學習得到的特徵資料作為所述有監督學習算法的自變量,以打標的歷史騙保人員作為因變量建立關係模型;
模型訓練模組1033,可以用於在所述關係模型的輸出達到預設準確率時確定構建的有監督學習算法。模型中參數的訓練迭代,在滿足輸出精度要求時可以作為線上使用。
本說明書實施例提供的伺服器或客戶端可以在電腦中由處理器執行相應的程式指令來實現,如使用windows操作系統的c++語言在PC端或伺服器端實現,或其他例如Linux、系統相對應的應用設計語言集合必要的硬體實現,或者基於量子電腦的處理邏輯實現等。因此,本說明書還提供一種保險欺詐識別的資料處理設備,具體的可以包括處理器以及用於儲存處理器可執行指令的記憶體,所述處理器執行所述指令時實現:
獲取待識別人群的關係關聯資料;
基於所述關係關聯資料構建所述待識別人群的多度關係網路圖資料以及提取所述待識別人群的人員特徵資料;
利用構建的有監督學習算法對所述待識別人群的多度關係網路圖資料和所述人員特徵資料進行識別,確所述待識別人群騙保輸出結果;所述有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。
上述的指令可以儲存在多種電腦可讀儲存媒體中。所述電腦可讀儲存媒體可以包括用於儲存資訊的物理裝置,可以將資訊數位化後再以利用電、磁或者光學等方式的媒體加以儲存。本實施例所述的電腦可讀儲存媒體有可以包括:利用電能方式儲存資訊的裝置如,各式記憶體,如RAM、ROM等;利用磁能方式儲存資訊的裝置如,硬碟、軟碟、磁帶、磁芯儲存器、磁泡儲存器、隨身碟;利用光學方式儲存資訊的裝置如,CD或DVD。當然,還有其他方式的可讀儲存媒體,例如量子記憶體、石墨烯記憶體等等。上述所述的裝置或伺服器或客戶端或處理設備中的所涉及的指令同上描述。
上述的處理設備可以具體的為保險伺服器或第三方服務機構提供保險反欺詐識別的伺服器,所述的伺服器可以為單獨的伺服器、伺服器集群、分布式系統伺服器或者處理設備請求資料的伺服器與其他相關聯資料處理的系統伺服器組合。因此,本說明書實施例還提供一種具體的伺服器產品,所述伺服器包括至少一個處理器以及用於儲存處理器可執行指令的記憶體,所述處理器執行所述指令時實現:
獲取待識別人群的關係關聯資料;
基於所述關係關聯資料構建所述待識別人群的多度關係網路圖資料以及提取所述待識別人群的人員特徵資料;
利用構建的有監督學習算法對所述待識別人群的多度關係網路圖資料和所述人員特徵資料進行識別,確所述待識別人群騙保輸出結果;所述有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。
需要說明的是,本說明書實施例上述所述的裝置和處理設備、伺服器,根據相關方法實施例的描述還可以包括其他的實施方式。具體的實現方式可以參照方法實施例的描述,在此不作一一贅述。
本說明書中的各個實施例均採用遞進的方式描述,各個實施例之間相同相似的部分互相參見即可,每個實施例重點說明的都是與其他實施例的不同之處。尤其,對於硬體+程式類實施例而言,由於其基本相似於方法實施例,所以描述的比較簡單,相關之處參見方法實施例的部分說明即可。
上述對本說明書特定實施例進行了描述。其它實施例在所附申請專利範圍的範圍內。在一些情況下,在申請專利範圍中記載的動作或步驟可以按照不同於實施例中的順序來執行並且仍然可以實現期望的結果。另外,在附圖中描繪的過程不一定要求示出的特定順序或者連續順序才能實現期望的結果。在某些實施方式中,多任務處理和並行處理也是可以的或者可能是有利的。
雖然本申請提供了如實施例或流程圖所述的方法操作步驟,但基於常規或者無創造性的勞動可以包括更多或者更少的操作步驟。實施例中列舉的步驟順序僅僅為眾多步驟執行順序中的一種方式,不代表唯一的執行順序。在實際中的裝置或系統伺服器產品執行時,可以按照實施例或者附圖所示的方法順序執行或者並行執行(例如並行處理器或者多執行緒處理的環境)。
儘管本說明書實施例內容中提到關係關聯資料的採集種類、訓練時選取的目標人群的範圍、判斷為騙保的概率計算方式等之類的資料獲取、儲存、互動、計算、判斷等操作和資料描述,但是,本說明書實施例並不局限於必須是符合行業通信標準、標準監督或無監督模型處理、通信協定和標準資料模型/模板或本說明書實施例所描述的情況。某些行業標準或者使用自定義方式或實施例描述的實施基礎上略加修改後的實施方案也可以實現上述實施例相同、等同或相近、或變形後可預料的實施效果。應用這些修改或變形後的資料獲取、儲存、判斷、處理方式等獲取的實施例,仍然可以屬於本說明書的可選實施方案範圍之內。
在20世紀90年代,對於一個技術的改進可以很明顯地區分是硬體上的改進(例如,對二極管、電晶體、開關等電路結構的改進)還是軟體上的改進(對於方法流程的改進)。然而,隨著技術的發展,當今的很多方法流程的改進已經可以視為硬體電路結構的直接改進。設計人員幾乎都透過將改進的方法流程編程到硬體電路中來得到相應的硬體電路結構。因此,不能說一個方法流程的改進就不能用硬體實體模組來實現。例如,可編程邏輯裝置(Programmable Logic Device, PLD)(例如現場可編程閘陣列(Field Programmable Gate Array,FPGA))就是這樣一種積體電路,其邏輯功能由用戶對裝置編程來確定。由設計人員自行編程來把一個數位系統“整合”在一片PLD上,而不需要請晶片製造廠商來設計和製作專用的積體電路晶片。而且,如今,取代手工地製作積體電路晶片,這種編程也多半改用“邏輯編譯器(logic compiler)”軟體來實現,它與程式開發撰寫時所用的軟體編譯器相類似,而要編譯之前的原始代碼也得用特定的編程語言來撰寫,此稱之為硬體描述語言(Hardware Description Language,HDL),而HDL也並非僅有一種,而是有許多種,如ABEL(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby Hardware Description Language)等,目前最普遍使用的是VHDL(Very-High-Speed Integrated Circuit Hardware Description Language)與Verilog。本領域技術人員也應該清楚,只需要將方法流程用上述幾種硬體描述語言稍作邏輯編程並編程到積體電路中,就可以很容易得到實現該邏輯方法流程的硬體電路。
控制器可以按任何適當的方式實現,例如,控制器可以採取例如微處理器或處理器以及儲存可由該(微)處理器執行的電腦可讀程式代碼(例如軟體或韌體)的電腦可讀媒體、邏輯閘、開關、專用積體電路(Application Specific Integrated Circuit,ASIC)、可編程邏輯控制器和嵌入微控制器的形式,控制器的例子包括但不限於以下微控制器:ARC 625D、Atmel AT91SAM、Microchip PIC18F26K20 以及Silicone Labs C8051F320,記憶體控制器還可以被實現為記憶體的控制邏輯的一部分。本領域技術人員也知道,除了以純電腦可讀程式代碼方式實現控制器以外,完全可以透過將方法步驟進行邏輯編程來使得控制器以邏輯閘、開關、專用積體電路、可編程邏輯控制器和嵌入微控制器等的形式來實現相同功能。因此這種控制器可以被認為是一種硬體部件,而對其內包括的用於實現各種功能的裝置也可以視為硬體部件內的結構。或者甚至,可以將用於實現各種功能的裝置視為既可以是實現方法的軟體模組又可以是硬體部件內的結構。
上述實施例闡明的處理設備、裝置、模組或單元,具體可以由電腦晶片或實體實現,或者由具有某種功能的產品來實現。一種典型的實現設備為電腦。具體的,電腦例如可以為個人電腦、膝上型電腦、車載人機互動設備、蜂巢式電話、相機電話、智慧電話、個人數位助理、媒體播放器、導航設備、電子郵件設備、遊戲控制台、平板電腦、可穿戴設備或者這些設備中的任何設備的組合。
雖然本說明書實施例提供了如實施例或流程圖所述的方法操作步驟,但基於常規或者無創造性的手段可以包括更多或者更少的操作步驟。實施例中列舉的步驟順序僅僅為眾多步驟執行順序中的一種方式,不代表唯一的執行順序。在實際中的裝置或終端產品執行時,可以按照實施例或者附圖所示的方法順序執行或者並行執行(例如並行處理器或者多執行緒處理的環境,甚至為分布式資料處理環境)。術語“包括”、“包含”或者其任何其他變體意在涵蓋非排他性的包含,從而使得包括一系列要素的過程、方法、產品或者設備不僅包括那些要素,而且還包括沒有明確列出的其他要素,或者是還包括為這種過程、方法、產品或者設備所固有的要素。在沒有更多限制的情況下,並不排除在包括所述要素的過程、方法、產品或者設備中還存在另外的相同或等同要素。
為了描述的方便,描述以上裝置時以功能分為各種模組分別描述。當然,在實施本說明書實施例時可以把各模組的功能在同一個或多個軟體及/或硬體中實現,也可以將實現同一功能的模組由多個子模組或子單元的組合實現等。以上所描述的裝置實施例僅僅是示意性的,例如,所述單元的劃分,僅僅為一種邏輯功能劃分,實際實現時可以有另外的劃分方式,例如多個單元或組件可以結合或者可以整合到另一個系統,或一些特徵可以忽略,或不執行。另一點,所顯示或討論的相互之間的耦合或直接耦合或通信連接可以是透過一些介面,裝置或單元的間接耦合或通信連接,可以是電性,機械或其它的形式。
本領域技術人員也知道,除了以純電腦可讀程式代碼方式實現控制器以外,完全可以透過將方法步驟進行邏輯編程來使得控制器以邏輯閘、開關、專用積體電路、可編程邏輯控制器和嵌入微控制器等的形式來實現相同功能。因此這種控制器可以被認為是一種硬體部件,而對其內部包括的用於實現各種功能的裝置也可以視為硬體部件內的結構。或者甚至,可以將用於實現各種功能的裝置視為既可以是實現方法的軟體模組又可以是硬體部件內的結構。
本發明是參照根據本發明實施例的方法、設備(系統)、和電腦程式產品的流程圖及/或方框圖來描述的。應理解可由電腦程式指令實現流程圖及/或方框圖中的每一流程及/或方框、以及流程圖及/或方框圖中的流程及/或方框的結合。可提供這些電腦程式指令到通用電腦、專用電腦、嵌入式處理機或其他可編程資料處理設備的處理器以產生一個機器,使得透過電腦或其他可編程資料處理設備的處理器執行的指令產生用於實現在流程圖一個流程或多個流程及/或方框圖一個方框或多個方框中指定的功能的裝置。
這些電腦程式指令也可儲存在能引導電腦或其他可編程資料處理設備以特定方式工作的電腦可讀記憶體中,使得儲存在該電腦可讀記憶體中的指令產生包括指令裝置的製造品,該指令裝置實現在流程圖一個流程或多個流程及/或方框圖一個方框或多個方框中指定的功能。
這些電腦程式指令也可裝載到電腦或其他可編程資料處理設備上,使得在電腦或其他可編程設備上執行一系列操作步驟以產生電腦實現的處理,從而在電腦或其他可編程設備上執行的指令提供用於實現在流程圖一個流程或多個流程及/或方框圖一個方框或多個方框中指定的功能的步驟。
在一個典型的配置中,計算設備包括一個或多個處理器(CPU)、輸入/輸出介面、網路介面和內存記憶體。
內存記憶體可能包括電腦可讀媒體中的非永久性記憶體,隨機存取記憶體(RAM)及/或非揮發性內存記憶體等形式,如唯讀記憶體(ROM)或閃存(flash RAM)。內存記憶體是電腦可讀媒體的示例。
電腦可讀媒體包括永久性和非永久性、可移除和非可移除媒體可以由任何方法或技術來實現資訊儲存。資訊可以是電腦可讀指令、資料結構、程式的模組或其他資料。電腦的儲存媒體的例子包括,但不限於相變內存記憶體(PRAM)、靜態隨機存取記憶體(SRAM)、動態隨機存取記憶體(DRAM)、其他類型的隨機存取記憶體(RAM)、唯讀記憶體(ROM)、電可抹除可編程唯讀記憶體(EEPROM)、快閃記憶體或其他內存記憶體技術、唯讀光碟唯讀記憶體(CD-ROM)、數位多功能光碟(DVD)或其他光學儲存、磁盒式磁帶,磁帶磁磁碟儲存或其他磁性儲存設備或任何其他非傳輸媒體,可用於儲存可以被計算設備存取的資訊。按照本文中的界定,電腦可讀媒體不包括暫存電腦可讀媒體(transitory media),如調變的資料信號和載波。
本領域技術人員應明白,本說明書的實施例可提供為方法、系統或電腦程式產品。因此,本說明書實施例可採用完全硬體實施例、完全軟體實施例或結合軟體和硬體方面的實施例的形式。而且,本說明書實施例可採用在一個或多個其中包含有電腦可用程式代碼的電腦可用儲存媒體(包括但不限於磁碟記憶體、CD-ROM、光學記憶體等)上實施的電腦程式產品的形式。
本說明書實施例可以在由電腦執行的電腦可執行指令的一般上下文中描述,例如程式模組。一般地,程式模組包括執行特定任務或實現特定抽象資料類型的例程、程式、物件、組件、資料結構等等。也可以在分布式計算環境中實踐本說明書實施例,在這些分布式計算環境中,由透過通信網路而被連接的遠端處理設備來執行任務。在分布式計算環境中,程式模組可以位於包括儲存設備在內的本地和遠端電腦儲存媒體中。
本說明書中的各個實施例均採用遞進的方式描述,各個實施例之間相同相似的部分互相參見即可,每個實施例重點說明的都是與其他實施例的不同之處。尤其,對於系統實施例而言,由於其基本相似於方法實施例,所以描述的比較簡單,相關之處參見方法實施例的部分說明即可。在本說明書的描述中,參考術語“一個實施例”、“一些實施例”、“示例”、“具體示例”、或“一些示例”等的描述意指結合該實施例或示例描述的具體特徵、結構、材料或者特點包含於本說明書實施例的至少一個實施例或示例中。在本說明書中,對上述術語的示意性表述不必須針對的是相同的實施例或示例。而且,描述的具體特徵、結構、材料或者特點可以在任一個或多個實施例或示例中以合適的方式結合。此外,在不相互矛盾的情況下,本領域的技術人員可以將本說明書中描述的不同實施例或示例以及不同實施例或示例的特徵進行結合和組合。
以上所述僅為本說明書實施例的實施例而已,並不用於限制本說明書實施例。對於本領域技術人員來說,本說明書實施例可以有各種更改和變化。凡在本說明書實施例的精神和原理之內所作的任何修改、等同替換、改進等,均應包含在本說明書實施例的申請專利範圍的範圍之內。
In order to enable those skilled in the art to better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present specification. Obviously, the described The examples are only a part of examples in this specification, but not all examples. Based on one or more embodiments in the present specification, all other embodiments obtained by a person having ordinary skill in the art without creative efforts should fall within the protection scope of the embodiments of the present specification.
Birds of a feather flock together. The scammers usually need multiple people to improve their camouflage. In many cases, the scam of insurance scammers is also based on acquaintance relationships or network characteristics that have obvious common characteristics or a certain dimension. For example, it is based on partnership scams by relatives, pyramid schemes with obvious stratified scam groups, and social or student groups drawn by leaders of experienced historical scams. The examples in this specification provide multiple implementations, triggering from a variety of relationship-related data of the target population including the insured and the claimant, to construct a multi-degree relationship network (the data of the relation network diagram can be called multi-degree relationship (Figure data), deeply digging the relationship network between the target groups, and solve the problem of low coverage and recognition rate that conventionally only identifies historical fraudsters and once-degree relationships that have a direct relationship with historical fraudsters. At the same time, the scheme provided by the embodiment of the present specification also takes into account the characteristics of the fraudsters themselves. For example, the fraudsters usually use false information to register an account, the account registration time is short, and the account holder mainly uses the insurance business after account registration. The implementation scheme provided in this specification combines historical relationship traits and self-characteristics of fraud insurance groups to mark out historical fraud officers and conduct algorithmic learning of a supervised model, so that it is possible to calculate or identify whether there are fraudulent insurance groups for people to be identified. result.
The following uses a specific application scenario of fraud identification processing for insurance services as an example to describe the implementation of this specification. Specifically, FIG. 1 is a schematic flowchart of an embodiment of a data processing method for identifying insurance fraud provided in this specification. Although this specification provides method operation steps or device structures as shown in the following embodiments or drawings, based on conventional or no creative labor, the method or device may include more or partly merged fewer operation steps. Or module unit. Among the steps or structures that do not logically have the necessary causal relationship, the execution order of these steps or the module structure of the device is not limited to the execution order or module structure shown in the embodiments of the present specification or the drawings. When the method or module structure is applied to an actual device, server, or end product, the method or module structure shown in the embodiment or the drawings may be executed sequentially or in parallel (for example, a parallel processor or Multi-threaded processing environment, even distributed processing, server cluster implementation environment).
Of course, the description of the following embodiments does not limit other technical solutions that can be extended based on this specification. For example, in other implementation scenarios, the implementation solutions provided in this specification can also be applied to the implementation scenarios of fund fraud identification, product transactions, and service transactions. A specific embodiment is shown in FIG. 1. A data processing method for insurance fraud identification provided in this specification may include:
S0: Obtain relationship data of the population to be identified;
S2: constructing a multiple degree relationship network diagram data of the crowd to be identified based on the relationship association data, and extracting personal characteristic data of the crowd to be identified;
S4: use the constructed supervised learning algorithm to identify the multiple degree relationship network graph data of the crowd to be identified and the personnel characteristic data, and confirm the fraudulent output of the crowd to be identified; the supervised learning algorithm includes The data relationship model is obtained by training with the selected target group's multi-relationship network data and personnel characteristics data, and the marked historical fraud insurance personnel as sample data. In the application scenario of this embodiment, usually insurance insurance, accounting, compensation and other links It is mainly aimed at claimants. In the examples of this specification, some actual scenarios are considered in which the scam motive exists from the beginning of the application. The main purpose of the scammer is to apply for the insurance payment amount. Of course, there are some after the application. Motivation for fraud. The insured is the main subject of the insurance, such as the insurance scam staff of a fellow villager intentionally creating an accident of the insured person. Therefore, in this embodiment, when identifying whether there is a target group at the time of fraud insurance, the claimant and the insured were selected. People collection. Therefore, in some embodiments of the method described in this specification, when a target group is selected for acquisition and learning of relationship characteristic data, the target group may include a set of personnel who apply for claims and the insured. It should be noted that in some implementation cases, the claimant may include an insurer, such as the father insuring the son, the father is the beneficiary, and the father is the claimant after the insurance; or in some implementations, the claimant may also include the insured Personnel, such as the insured who insures himself, the beneficiary is himself. The claimants and insured persons described above can understand the names of the categories of persons in different roles in the insurance business, and are not currently different persons. The claimant and insured persons shown in some implementation scenarios can All or part of the same.
Of course, in other embodiments, the target group may also be selected by selecting one or more of a claim applicant, an insurer, an insured or a beneficiary.
The relationship-related data may include multiple dimensions of data related to the people in the target group, such as household registration, age, kinship / classmate relations between people, insurance data, insurance outflow data, and so on. The specific relationship related data can be selected according to the actual application scenario to determine which types of which data are used. Generally, the operator can use the data information that may be involved in the insurance fraud as the basis for collecting the relationship related data. In an embodiment provided in this specification, the relationship related data may include at least one of the following:
Social relationship data, terminal data, terminal applications and application account operation information, behavior data related to insurance behavior, personnel basic attribute data, geographic location data.
The social relationship data may include social relationships among persons in the target group, such as cousins, teachers and students, family members, classmates, leaders and subordinates. The terminal information may include the brand, model, and type of communication equipment used by personnel. In some fraud scenarios, personnel use mobile phones of the same brand. The terminal application and application account operation information can be used to determine whether to use the same application and use the same account to log in to different terminals to perform insurance fraud operations. In some scenarios, many of the following are subject to the command of the boss to be directed on the terminal application. Do it. The behavior data related to the insurance behavior may include behavior data of the target person's insurance behavior, claim settlement behavior, compensation amount, etc. The personnel basic attribute data may include the age, gender, occupation, and household registration of the policyholder / applicant. The geographic location data may include information on the geographic location where the target population is currently located or information on areas that have historically passed / retained fruit. Of course, the data relationship related data of each dimension described above may also have other definitions or contain more / less data categories and information, and may also include relationship related data of other dimensions other than the above, such as consumer information Even credit information or administrative penalty information, specific collection of one or more of the above information can be collected.
The personnel characteristic data may include information related to a single person, such as gender, age, insurance service account or terminal application account registration time, credit history, consumption situation, etc., or may also include behaviors associated with insurance behaviors. Information, such as repeated insurance acts, regular claims, and whether the amount of compensation is normal. It can also include the following transaction information of other goods or services, such as long-term large expenditures, multiple auto insurances, purchase of multiple mobile phones, registration of multiple communication accounts / social accounts, etc.
The person characteristic data used in the calculation of specific person characteristics may adopt one or a combination of the above to realize the identification of the person's own characteristics. Therefore, in another embodiment, the personnel characteristic data may include characteristic data extracted from at least one of a user registration account, transaction data, and behavior data associated with insurance behavior.
There is usually a relatively close relationship network among the fraudsters. In this embodiment, the obtained multi-dimensional relationship association data can be used to construct multiple degree relationship network graph data of the target group. The multi-degree relationship network diagram data may include a relationship network diagram generated based on a relationship chain between different persons established by the relationship association data, where the relationship chain data between persons on the relationship network diagram is Multiple degree network graph data. The relationship chain can represent the relationship data between every two persons, such as A and B being the boss relationship and A and C being the family relationship. The relationship between two separate persons may be referred to as a one-degree relationship, and the “multi-degree” in the multi-degree relationship network diagram data described in this embodiment may include a new person established based on the one-degree relationship. Relevant data, such as the second-degree relationship between the first person and the third person based on the first-degree relationship between the first person and the second person and the first-degree relationship between the second person and the third person, or even further based on other first-degree relationships The third relationship between the first person and the fourth person, and so on.
For example, if A is a single person and B is A's brother-in-law, then A and B are once a social relationship. There was no social relationship between A and his brother-in-law's company owner C, but in the embodiment of this specification, because B is both A's brother-in-law and the subordinate of company owner C, so the second-degree relationship between A and company owner C is established.
In addition to the social relationships between the above-mentioned people, other types of multidegree relationship network diagram data can be formed according to the adopted relationship association data or relationship construction requirements, such as whether they are natives, using the same communication tool, and multi-person terminals. Of an app is logged in at a fixed time, etc. Of course, in the specific implementation of constructing a relationship network based on the relationship association data, the determination between the relationships may be designed in advance to establish rules for the relationship chain.
Based on the well-established multi-degree relational network graph data and the extracted personnel characteristic data, this embodiment can use a supervised learning algorithm to learn the relationship characteristics and self-characteristics of the fraudsters, so that an effective identification model can be established.
Generally, the commonly used methods of machine learning are mainly divided into supervised learning, sometimes also referred to as supervised learning and unsupervised learning. Supervised learning is a classification processing method. Usually for labeled data sets, an optimal model is trained through existing training samples (that is, known data and its corresponding output) (this model belongs to a set of functions, The best means that it is the best under a certain evaluation criterion), and then use this model to map all the inputs to the corresponding output, make a simple judgment on the output to achieve the purpose of classification, and also have the ability to perform the analysis of unknown data. Classification ability. Typical examples in supervised learning are KNN (k-NearestNeighbor, neighborhood algorithm), SVM (Support Vector Machine, Support Vector Machine). With a certain number of training samples, supervised learning algorithms can obtain more accurate output results than unsupervised algorithms.
According to the different supervised learning algorithms selected, other specific relational features and their own characteristics are designed and determined according to the type of algorithm and recognition processing requirements. For example, a supervised graph algorithm such as Structure2vec can be used. For example, in one embodiment, the constructed supervised learning algorithm includes:
S40: Use the selected supervised learning algorithm to perform a first relational network learning on the relationship characteristics of the target person and other personnel in the multi-relationship network data of the target population, and perform a second self based on the self-characteristic data of the target person characteristics. Attribute learning
S42: Use the feature data obtained by the first relationship network learning and the second self attribute learning as the independent variables of the supervised learning algorithm, and use the marked historical fraud insurance personnel as the dependent variable to establish the relationship model;
S44: Determine the constructed supervised learning algorithm when the output of the relational model reaches a preset accuracy rate.
FIG. 2 is a schematic diagram of a process of constructing an embodiment of a supervised learning algorithm provided in the present specification.
In the example shown in Figure 2, Structured 2vec's supervised graph algorithm can be used: on the one hand, to learn the relationship characteristics of the target person and its neighbors (such as how many people are related, whether they are related to fraudsters), on the other hand Learn the characteristics of the target person (such as gender, age, etc.), the above characteristics are used as the x variable of the model; secondly, whether the insurance staff are marked according to the history as the y variable; finally, the relevant model is established according to y and x to achieve Only x can predict y.
In the application scenario of this embodiment, it may be a single person who finally identifies whether it is fraud insurance. That is, the reason in this embodiment is that the supervised learning algorithm learns the relationship characteristics of gang fraud, and then combines the characteristics of the scammers themselves to directly obtain the probability of whether a certain person to be identified is a scammer or a scammer. Scam output. For example, the probability that a person can be marked as a fraudulent or normal person, or as a fraudulent person.
Of course, the tags labeled as fraudsters here are based on the recognition results of relationship characteristics and their own characteristics, which can be used as the basis and reference for determining whether these persons are fraudsters. The final determination of whether it is fraudulent insurance can be subjective judgment of the operator, or combined with other calculation methods for judgment and determination.
The data processing method for insurance fraud provided by this embodiment can construct a multi-degree relationship network graph data of a population based on multi-dimensional relationship association data of an insured person and an insured person, and can further dig out the relationship network between personnel To improve recognition efficiency and scope. At the same time, combined with the characteristics of the fraudsters themselves, a supervised learning model is jointly established to learn the characteristics of the relationship network and the characteristics of the fraudsters. Gang fraudsters not only have obvious and abundant relationship characteristics on the relationship network, but their own characteristics often also show similarities. Therefore, the methods provided in the examples of this specification can more effectively and efficiently identify fraudsters To improve the efficiency of recognition processing.
In another embodiment of the method provided in the present specification, the information of historical fraudsters can also be used in combination with the multi-degree relationship network map data to identify fraudsters. Specifically, in another embodiment of the method provided in this specification, the relationship-related data may further include: historical fraudulent personnel list data.
In this embodiment, the information of historical fraudsters is added. When analyzing and processing the classified community, the degree of participation of historical fraudsters is considered. In general, if the historical fraudsters have a higher concentration in a certain classification community, the more likely the persons in the classification community to perform fraud protection. The concentration of the relationship described in this embodiment may include the degree of participation of historical scammers, and specifically may include the number of historical scammers in the classification community, the proportion of historical scammers, historical scammers and other personnel Close relationship. An example of the intensity of the relationship is described. In a 10-person risk community, 2 historical insurance scammers have one-degree or more-degree kinship with the other 6 persons, and 2 persons have class relationships. Indicate that it may be a pyramid scheme scam. The specific relationship concentration can be calculated in different ways, such as the number of historical insurance deceivers, the proportion, and the relationship network. In another embodiment of the present specification, the relationship concentration may be calculated from two indicators: the size of the population to be identified and the number of historical fraudsters. The relationship density may be used as a value for measuring the probability of fraudulent insurance. Specifically, it can include:
Taking the logarithm of the number of people in the crowd to be identified as the first factor;
Taking the proportion of the number of historical fraud insurance personnel among the persons to be identified as the second factor;
Based on the product of the first factor and the second factor as the group fraud insurance probability of the population to be identified.
Then, the value of individual fraud insurance probabilities calculated based on its own characteristics can be calculated with the group fraud insurance probabilities to determine the probability that the final output group is fraud insurance or a single person is fraud insurance. Alternatively, the group fraud insurance probability and the individual fraud insurance probability are used separately, and no mutual calculation is performed.
For example, in specific implementation, the following methods can be used to calculate the probability of community fraud:
RiskDegree = log (total number of people in the classification community) * number of historical fraudsters / total number of people in the classification community.
Of course, other calculation methods or deformation and transformation methods may also be adopted, such as taking natural logarithms, which are limited and detailed here.
The foregoing embodiment provides the information that can be used to identify fraudulent fraud groups by using historical information of fraudulent insurance personnel. In another embodiment provided in this specification, the characteristics of the relationship network between various members of the crowd can be used to determine whether it is a fraudster. Specifically, such as determining the network structure characteristics of the relationship between people in the crowd;
If the characteristics of the network structure conform to the default fraud network structure, the crowd is marked as a fraud group.
The above-mentioned manner can be used in the training of a supervised learning algorithm, and the crowd is a target person. In the process of identifying a person to be identified, the crowd is the person to be identified.
The network structure characteristics may be based on the information of the people in the crowd, the network information of the relationships among the people, and the like. The relationship network information here may be the aforementioned one-degree information, or may include the constructed multi-degree information.
A certain algorithm can be used to identify and analyze the characteristics of the relationship network in the community. If the network structure characteristics meet the characteristics of the fraud protection group, it can be marked as a fraud group at this time. For example, in an example, the relationship network in the crowd may be a network structure such as a "spherical network", a "pyramid network", and the like. The "pyramid network" is similar to a MLM organization. It has a layer-by-layer relationship structure, which is more likely to belong to scams. The "spherical network" is a network that is related to each other and may be a decentralized scam.
An embodiment of the present specification provides a method for data processing of insurance fraud, using relational data close to the actual relational network to support the mining of relational network algorithms, and the calculation of relational network data with multiple relations. Based on the multi-dimensional relationship data of the insured and the insured, constructing a multi-degree relationship network graph of the crowd can further dig the relationship network between the personnel and improve the identification efficiency and scope. At the same time, combined with the characteristics of the fraudsters themselves, a supervised learning model is jointly established to learn the characteristics of the relationship network and the characteristics of the fraudsters. Gang fraudsters not only have obvious and abundant relationship characteristics on the relationship network, but their own characteristics often also show similarities. Therefore, the methods provided in the examples of this specification can more effectively and efficiently identify fraudsters To improve the efficiency of recognition processing.
The method described above can be used to identify insurance fraud on the client side, such as installing anti-fraud applications and payment applications on mobile terminals for insurance services. The client may be a personal computer (PC), a server, an industrial control computer (industrial control computer), a mobile smart phone, a tablet electronic device, a portable computer (such as a notebook computer), a personal digital assistant (PDA), or Desktop computers or smart wearables. Mobile communication terminals, handheld devices, in-vehicle devices, wearable devices, television devices, computing devices. It can also be applied to the system server of the insurance business party or service party or a third party organization. The system server may include a separate server, a server cluster, a distributed system server, or a server that processes data requested by the equipment. In combination with other associated data processing system servers.
The method embodiments provided in the embodiments of this specification may be executed in a mobile terminal, a computer terminal, a server, or a similar computing device. Taking the server as an example, FIG. 3 is a block diagram of a hardware structure of a server for identifying a damaged part of a vehicle according to an embodiment of the present invention. As shown in FIG. 3, the server 10 may include one or more (only one shown in the figure) a processor 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) , A memory 104 for storing data, and a transmission module 106 for communication functions. Persons of ordinary skill in the art can understand that the structure shown in FIG. 3 is only schematic, and it does not limit the structure of the electronic device. For example, the server 10 may further include more or fewer components than those shown in FIG. 3, and may further include other processing hardware, such as a database or a multi-level cache, or have a configuration different from that shown in FIG. 3. .
The memory 104 may be used to store software programs and modules of application software, such as program instructions / modules corresponding to the search method in the embodiment of the present invention. The processor 102 runs the software programs and modules stored in the memory 104. Therefore, various functional applications and data processing are performed, that is, the processing method for displaying the content of the navigation interactive interface described above is implemented. The memory 104 may include high-speed random memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely disposed relative to the processor 102, and these remote memories may be connected to the computer terminal 10 through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
The transmission module 106 is used to receive or send data through a network. Specific examples of the above-mentioned network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network adapter (NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission module 106 may be a radio frequency (RF) module, which is used to communicate with the Internet in a wireless manner.
Based on the device model identification method described above, this specification also provides a data processing device for insurance fraud identification. The device may include a system (including a distributed system), software (application), a module, a component, a server, a client, and the like that use the method described in the embodiment of the present specification, and a device device that combines necessary implementation hardware . Based on the same innovative concept, the processing device in one embodiment provided in this specification is as described in the following embodiments. Since the implementation solution of the device to solve the problem is similar to the method, the implementation of the specific processing device in the embodiment of this specification may refer to the implementation of the foregoing method, and the duplicated details are not described again. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and conceived. Specifically, as shown in FIG. 4, FIG. 4 is a schematic diagram of a module structure of an embodiment of a data processing device for identifying insurance fraud provided in this specification, which may include:
The data acquisition module 101 may be used to acquire relationship related data of a population to be identified;
The feature calculation module 102 may be configured to construct a multiple degree relationship network graph data of the population to be identified based on the relationship association data, and extract person characteristic data of the population to be identified;
The fraud identification module 103 may be used to identify the multiple degree relationship network graph data of the crowd to be identified and the characteristics of the person using the constructed supervised learning algorithm, to confirm the fraudulent output of the crowd to be identified; The supervised learning algorithm includes a data relationship model that is trained by using the multi-relational network data and personnel characteristic data of the selected target population, and the marked historical fraud insurance personnel as sample data.
In a specific embodiment of the device, the relationship-related data may include at least one of the following:
Social relationship data, terminal data, terminal applications and application account operation information, behavior data related to insurance behavior, personnel basic attribute data, geographic location data.
In another embodiment of the device, the fraud recognition module 103 determines that the fraudulent output result of the to-be-identified group includes outputting a probability of whether a single to-be-identified target person is a fraudulent person or a fraudulent person.
In another embodiment of the device, the selected target group includes a set of persons applying for claims and the insured.
In another embodiment of the device, the personnel characteristic data includes characteristic data extracted from at least one of a user registration account, transaction data, and behavior data associated with an insurance behavior.
FIG. 5 is another embodiment of the device. As shown in FIG. 5, the fraud identification module 103 includes:
A feature learning module 1031 may be used to perform first-relation network learning on the relationship characteristics of the target person and other persons in the multi-relationship network data of the target population using the selected supervised learning algorithm, based on the characteristics of the target person. Self-characteristic data for second self-attribute learning;
The relationship establishing module 1032 may be used to use the feature data obtained by the first relationship network learning and the second self attribute learning as the independent variables of the supervised learning algorithm, and use the marked historical fraud insurance personnel as the dependent variable to establish the relationship. model;
The model training module 1033 may be used to determine a supervised learning algorithm to be constructed when the output of the relational model reaches a preset accuracy rate. The training iterations of the parameters in the model can be used online when the output accuracy requirements are met.
The server or client provided in the embodiments of this specification can be implemented by the processor executing corresponding program instructions in the computer, such as using the C ++ language of the Windows operating system on the PC or server, or other systems such as Linux and the system. Corresponding application design language sets the necessary hardware implementation, or the processing logic implementation based on quantum computers. Therefore, this specification also provides a data processing device for identifying insurance fraud, which may specifically include a processor and a memory for storing processor-executable instructions. When the processor executes the instructions, the processor implements:
Obtain relationship data of the population to be identified;
Constructing a multi-degree relationship network diagram data of the crowd to be identified based on the relationship association data and extracting person characteristic data of the crowd to be identified;
The constructed supervised learning algorithm is used to identify the multiple degree network graph data of the crowd to be identified and the characteristics of the person, and to confirm the fraudulent output of the crowd; the supervised learning algorithm includes using The data relationship model of the selected target population based on the multi-relationship network data and personnel characteristics data, and the marked historical fraud insurance personnel are trained as sample data.
The above instructions can be stored in a variety of computer-readable storage media. The computer-readable storage medium may include a physical device for storing information, and the information may be digitized and then stored by using a medium such as electricity, magnetism, or optics. The computer-readable storage medium described in this embodiment may include: a device for storing information using electric energy, such as various types of memory, such as RAM, ROM, etc .; a device for storing information using magnetic energy, such as hard disk, floppy disk, Magnetic tape, magnetic core storage, bubble storage, flash drives; devices that use optical means to store information, such as CDs or DVDs. Of course, there are other ways of readable storage media, such as quantum memory, graphene memory, and so on. The instructions involved in the device or server or client or processing device described above are the same as described above.
The above processing equipment may specifically provide an insurance anti-fraud identification server for an insurance server or a third-party service agency, and the server may be a separate server, a server cluster, a distributed system server, or a processing equipment request The data server is combined with other system servers associated with data processing. Therefore, the embodiment of the present specification also provides a specific server product. The server includes at least one processor and a memory for storing processor-executable instructions. When the processor executes the instructions, it implements:
Obtain relationship data of the population to be identified;
Constructing a multi-degree relationship network diagram data of the crowd to be identified based on the relationship association data and extracting person characteristic data of the crowd to be identified;
The constructed supervised learning algorithm is used to identify the multiple degree network graph data of the crowd to be identified and the characteristics of the person, and to confirm the fraudulent output of the crowd; the supervised learning algorithm includes using The data relationship model of the selected target population based on the multi-relationship network data and personnel characteristics data, and the marked historical fraud insurance personnel are trained as sample data.
It should be noted that the devices, processing equipment, and servers described above in the embodiments of this specification may also include other implementations according to the description of the related method embodiments. For specific implementation manners, reference may be made to the description of the method embodiments, and details are not described herein.
Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the hardware + programming embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts may refer to the description of the method embodiment.
The specific embodiments of the present specification have been described above. Other embodiments are within the scope of the appended patent applications. In some cases, the actions or steps described in the scope of the patent application may be performed in a different order than in the embodiments and still achieve the desired result. In addition, the processes depicted in the figures do not necessarily require the particular order shown or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
Although the present application provides method operation steps as described in the embodiments or flowcharts, more or less operation steps may be included based on conventional or non-creative labor. The sequence of steps listed in the embodiments is only one way of executing the steps, and does not represent the only sequence of execution. When the actual device or system server product is executed, it may be executed sequentially or in parallel according to the method shown in the embodiment or the drawings (for example, a parallel processor or a multi-threaded processing environment).
Although the content of the examples in this specification mentions the types of collection of relationship-related data, the range of target groups selected during training, the probability calculation method for judging scams, and other operations such as data acquisition, storage, interaction, calculation, and judgment, and The data description, however, the embodiments of the present specification are not limited to the situations that must conform to industry communication standards, standard supervision or unsupervised model processing, communication protocols and standard data models / templates, or the embodiments described in this specification. Certain industry standards or implementations that are slightly modified based on implementations described in custom methods or embodiments can also achieve the same, equivalent or similar, or predictable implementation effects of the above embodiments. Embodiments obtained by applying these modified or deformed data acquisition, storage, judgment, processing methods, etc., may still fall within the scope of optional implementations of this specification.
In the 1990s, for a technical improvement, it can be clearly distinguished whether it is an improvement in hardware (for example, the improvement of circuit structures such as diodes, transistors, switches, etc.) or an improvement in software (for the improvement of method flow). . However, with the development of technology, the improvement of many methods and processes can be regarded as a direct improvement of the hardware circuit structure. Designers almost always get the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be realized by a hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. It is programmed by the designer to "integrate" a digital system on a PLD, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of making integrated circuit chips by hand, this programming is mostly implemented using "logic compiler" software, which is similar to the software compiler used in program development and writing, and requires compilation. The previous original code must also be written in a specific programming language. This is called the Hardware Description Language (HDL). There is not only one kind of HDL, but many types, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog are commonly used. Those skilled in the art should also be clear that the hardware circuit that implements the logic method flow can be easily obtained by simply programming the method flow into the integrated circuit with the above-mentioned several hardware description languages.
The controller may be implemented in any suitable manner, for example, the controller may take the form of a microprocessor or processor and a computer-readable storage of computer-readable program code (such as software or firmware) executable by the (micro) processor. Media, logic gates, switches, application specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that, in addition to implementing the controller in pure computer-readable program code, it is entirely possible to make the controller use logic gates, switches, dedicated integrated circuits, programmable logic controllers, and Embedded in the form of a microcontroller, etc. to achieve the same function. Therefore, the controller can be considered as a hardware component, and the device included in the controller for implementing various functions can also be considered as a structure in the hardware component. Or even, a device for implementing various functions can be regarded as a structure that can be both a software module implementing the method and a hardware component.
The processing equipment, devices, modules, or units described in the above embodiments may be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a vehicle-mounted human-machine interactive device, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, Tablet, wearable, or a combination of any of these.
Although the embodiments of the present specification provide the operation steps of the method as described in the embodiments or flowcharts, more or less operation steps may be included based on conventional or non-creative means. The sequence of steps listed in the embodiments is only one way of executing the steps, and does not represent the only sequence of execution. When the actual device or terminal product is executed, it may be executed sequentially or in parallel according to the method shown in the embodiment or the accompanying drawings (for example, a parallel processor or multi-threaded processing environment, or even a distributed data processing environment). The terms "including,""including," or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, product, or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed Elements, or elements that are inherent to such a process, method, product, or device. Without further limitation, it does not exclude that there are other identical or equivalent elements in the process, method, product or equipment including the elements.
For the convenience of description, when describing the above device, the functions are divided into various modules and described separately. Of course, when implementing the embodiments of this specification, the functions of each module may be implemented in the same or multiple software and / or hardware, or the module that implements the same function may be composed of multiple submodules or subunits. Implementation etc. The device embodiments described above are only schematic. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner. For example, multiple units or components may be combined or integrated into Another system, or some features, can be ignored or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, which may be electrical, mechanical or other forms.
Those skilled in the art also know that, in addition to implementing the controller in pure computer-readable program code, it is entirely possible to make the controller use logic gates, switches, dedicated integrated circuits, programmable logic controllers, and Embedded in the form of a microcontroller, etc. to achieve the same function. Therefore, such a controller can be considered as a hardware component, and the device included in the controller for implementing various functions can also be considered as a structure within the hardware component. Or even, a device for implementing various functions can be regarded as a structure that can be both a software module implementing the method and a hardware component.
The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and a combination of the process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to generate a machine, so that instructions generated by the processor of the computer or other programmable data processing device can be used to generate instructions. Means for realizing the functions specified in one or more flowcharts and / or one or more blocks of the block diagrams.
These computer program instructions may also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including a command device, The instruction device implements a function specified in a flowchart or a plurality of processes and / or a block or a block of the block diagram.
These computer program instructions can also be loaded on a computer or other programmable data processing device, so that a series of operating steps can be performed on the computer or other programmable device to generate a computer-implemented process, which can be executed on the computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more flowcharts and / or one or more blocks of the block diagrams.
In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.
Memory memory may include non-permanent memory, random access memory (RAM), and / or non-volatile memory in computer-readable media, such as read-only memory (ROM) or flash memory (flash RAM) ). Memory is an example of a computer-readable medium.
Computer-readable media includes permanent and non-permanent, removable and non-removable media. Information can be stored by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), and other types of random access memory (RAM ), Read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory-memory technologies, CD-ROM, CD-ROM A functional optical disk (DVD) or other optical storage, magnetic tape cartridge, magnetic disk storage or other magnetic storage device or any other non-transmitting medium may be used to store information that can be accessed by a computing device. According to the definition in this article, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.
Those skilled in the art should understand that the embodiments of the present specification may be provided as a method, a system, or a computer program product. Therefore, the embodiments of the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present specification may use a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk memory, CD-ROM, optical memory, etc.) containing computer-usable program codes. form.
The embodiments of this specification can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The embodiments of the present specification can also be practiced in distributed computing environments. In these distributed computing environments, tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.
Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant part, refer to the description of the method embodiment. In the description of this specification, the description with reference to the terms “one embodiment”, “some embodiments”, “examples”, “specific examples”, or “some examples” and the like means specific features described in conjunction with the embodiments or examples , Structure, materials, or features are included in at least one embodiment or example of an embodiment of the present specification. In this specification, the schematic expressions of the above terms are not necessarily directed to the same embodiment or example. Moreover, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. In addition, without any contradiction, those skilled in the art may combine and combine different embodiments or examples and features of the different embodiments or examples described in this specification.
The above descriptions are merely examples of the embodiments of the present specification, and are not intended to limit the embodiments of the present specification. For those skilled in the art, the embodiments of the present specification may have various modifications and changes. Any modification, equivalent replacement, and improvement made within the spirit and principle of the embodiments of the present specification shall be included in the scope of patent application of the embodiments of the present specification.

10‧‧‧伺服器10‧‧‧Server

102‧‧‧處理器 102‧‧‧ processor

104‧‧‧記憶體 104‧‧‧Memory

106‧‧‧傳輸模組 106‧‧‧Transmission Module

101‧‧‧資料獲取模組 101‧‧‧Data Acquisition Module

102‧‧‧特徵計算模組 102‧‧‧Feature Calculation Module

103‧‧‧欺詐識別模組 103‧‧‧ Fraud Identification Module

1031‧‧‧特徵學習模組 1031‧‧‧Feature Learning Module

1032‧‧‧關係建立模組 1032‧‧‧ Relationship Building Module

1033‧‧‧模型訓練模組 1033‧‧‧Model Training Module

為了更清楚地說明本說明書實施例或現有技術中的技術方案,下面將對實施例或現有技術描述中所需要使用的附圖作簡單地介紹,顯而易見地,下面描述中的附圖僅僅是本說明書中記載的一些實施例,對於本領域普通技術人員來講,在不付出創造性勞動性的前提下,還可以根據這些附圖獲得其他的附圖。In order to more clearly explain the embodiments of the present specification or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only For some ordinary people skilled in the art, some embodiments described in the description can also obtain other drawings according to the drawings without paying creative labor.

圖1是本說明書提供的一種保險欺詐識別的資料處理方法實施例的流程示意圖; FIG. 1 is a schematic flowchart of an embodiment of a data processing method for identifying insurance fraud provided in this specification; FIG.

圖2是本說明書提供的一種構建有監督識別模型的處理過程示意圖; FIG. 2 is a schematic diagram of a process of constructing a supervised identification model provided in the present specification; FIG.

圖3是本說明書提供的一種保險欺詐識別處理伺服器的硬體結構框圖; FIG. 3 is a block diagram of a hardware structure of an insurance fraud identification processing server provided in this specification; FIG.

圖4是本說明書提供的一種保險欺詐識別的資料處理裝置的模組結構示意圖。 FIG. 4 is a schematic diagram of a module structure of a data processing device for insurance fraud identification provided in the present specification.

圖5是本說明書提供的一種保險欺詐識別的資料處理裝置中欺詐識別模組的模組結構示意圖。 FIG. 5 is a schematic diagram of a module structure of a fraud identification module in a data processing device for insurance fraud identification provided in the present specification.

Claims (14)

一種保險欺詐識別的資料處理方法,該方法包括: 獲取待識別人群的關係關聯資料; 基於該關係關聯資料構建該待識別人群的多度關係網路圖資料以及提取該待識別人群的人員特徵資料; 利用構建的有監督學習算法對該待識別人群的多度關係網路圖資料和該人員特徵資料進行識別,確該待識別人群騙保輸出結果;該有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。A data processing method for identifying insurance fraud, the method includes: Obtain relationship data of the population to be identified; Constructing a multi-degree relationship network diagram data of the crowd to be identified based on the relationship association data and extracting personal characteristic data of the crowd to be identified; The constructed supervised learning algorithm is used to identify the multi-relationship network graph data of the crowd to be identified and the characteristics of the person to confirm the fraudulent output of the crowd; the supervised learning algorithm includes the selected target population The multi-relationship network data and personnel characteristics data, the data relationship model obtained by training historical fraud insurance personnel as sample data. 如申請專利範圍第1項所述的方法,該關係關聯資料包括下述中的至少一種: 社會關係資料、終端資料、終端的應用以及應用帳戶操作資訊、與保險行為關聯的行為資料、人員基礎屬性資料、地理位置資料。According to the method described in the first patent application scope, the relationship-related data includes at least one of the following: Social relationship data, terminal data, terminal applications and application account operation information, behavior data related to insurance behavior, personnel basic attribute data, geographic location data. 如申請專利範圍第1項所述的方法,所述確該待識別人群騙保輸出結果包括輸出單個待識別目標人員是否為欺詐人員或為欺詐人員的概率。According to the method described in item 1 of the scope of the patent application, the output of confirming the fraudulent protection of the group to be identified includes outputting the probability of whether a single target person to be identified is a fraudulent person or a fraudulent person. 如申請專利範圍第1項所述的方法,該選取的目標人群包括申請理賠人員和被保險人的人員集合。According to the method described in the first scope of the patent application, the selected target group includes a set of persons applying for claims and the insured. 如申請專利範圍第1或3項中任意一項所述的方法, 該人員特徵資料包括用戶註冊帳號、交易資料、與保險行為關聯的行為資料中的至少一項提取出來的特徵資料。According to the method described in any one of items 1 or 3 of the scope of patent application, the personnel characteristic data includes characteristic data extracted from at least one of a user registration account, transaction data, and behavior data associated with insurance behavior. 如申請專利範圍第1或3項中任意一項所述的方法,該採用下述方式構建有監督學習算法包括: 利用選取的有監督學習算法對目標人群的多度關係網路資料中目標人員與其他人員的關係特徵進行第一關係網路學習、基於該目標人員特徵的自身特徵資料進行第二自身屬性學習; 以該第一關係網學習和第二自身屬性學習得到的特徵資料作為該有監督學習算法的自變量,以打標的歷史騙保人員作為因變量建立關係模型; 在該關係模型的輸出達到預設準確率時確定構建的有監督學習算法。According to the method described in any one of claims 1 or 3, the method of constructing a supervised learning algorithm in the following ways includes: Use the selected supervised learning algorithm to perform the first relational network learning on the relationship characteristics of the target person and other personnel in the multi-relationship network data of the target group, and the second self attribute learning based on the self-characteristic data of the target person's characteristics; Use the feature data obtained from the first relationship network learning and the second self-attribute learning as the independent variables of the supervised learning algorithm, and use the marked historical fraud insurance personnel as the dependent variable to establish the relationship model; A supervised learning algorithm is determined when the output of the relational model reaches a preset accuracy rate. 一種保險欺詐識別的資料處理裝置,包括: 資料獲取模組,用於獲取待識別人群的關係關聯資料; 特徵計算模組,用於基於該關係關聯資料構建該待識別人群的多度關係網路圖資料以及提取該待識別人群的人員特徵資料; 欺詐識別模組,用於利用構建的有監督學習算法對該待識別人群的多度關係網路圖資料和該人員特徵資料進行識別,確該待識別人群騙保輸出結果;該有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。A data processing device for identifying insurance fraud includes: A data acquisition module for acquiring relationship data of a population to be identified; A feature calculation module, which is used to construct the abundance relationship network graph data of the group to be identified based on the relationship association data and extract the person characteristic data of the group to be identified; A fraud recognition module is used to identify the multi-relationship network graph data of the crowd to be identified and the characteristics of the person using the constructed supervised learning algorithm, and to confirm the fraudulent output of the crowd; Including the data relationship model obtained by training with the selected target group's multiple-relationship network data and personnel characteristics data, and marked historical fraud insurance personnel as sample data. 如申請專利範圍第7項所述的裝置,其中,該關係關聯資料包括下述中的至少一種: 社會關係資料、終端資料、終端的應用以及應用帳戶操作資訊、與保險行為關聯的行為資料、人員基礎屬性資料、地理位置資料。The device according to item 7 of the scope of patent application, wherein the relationship-related data includes at least one of the following: Social relationship data, terminal data, terminal applications and application account operation information, behavior data related to insurance behavior, personnel basic attribute data, geographic location data. 如申請專利範圍第7項所述的裝置,該欺詐識別模組確該待識別人群騙保輸出結果包括輸出單個待識別目標人員是否為欺詐人員或為欺詐人員的概率。According to the device described in claim 7 of the patent application scope, the fraud identification module confirms that the fraudulent output of the to-be-identified group includes the probability of outputting whether a single to-be-identified target person is a fraudulent person or a fraudulent person. 如申請專利範圍第7項所述的裝置,該選取的目標人群包括申請理賠人員和被保險人的人員集合。As for the device described in the seventh item of the scope of patent application, the selected target group includes a group of persons applying for claims and the insured. 如申請專利範圍第7或9項所述的裝置,該人員特徵資料包括用戶註冊帳號、交易資料、與保險行為關聯的行為資料中的至少一項提取出來的特徵資料。As for the device described in item 7 or 9 of the scope of patent application, the personnel characteristic data includes characteristic data extracted from at least one of a user registration account, transaction data, and behavior data associated with insurance behavior. 如申請專利範圍第7或9項所述的裝置,該欺詐識別模組包括: 特徵學習模組,用於利用選取的有監督學習算法對目標人群的多度關係網路資料中目標人員與其他人員的關係特徵進行第一關係網路學習、基於該目標人員特徵的自身特徵資料進行第二自身屬性學習; 關係建立模組,用於以該第一關係網學習和第二自身屬性學習得到的特徵資料作為該有監督學習算法的自變量,以打標的歷史騙保人員作為因變量建立關係模型; 模型訓練模組,用於在該關係模型的輸出達到預設準確率時確定構建的有監督學習算法。According to the device described in claim 7 or 9, the fraud identification module includes: Feature learning module, used to use the selected supervised learning algorithm to perform first-relationship network learning on the relationship characteristics between the target person and other personnel in the target's abundance relationship network data, based on the characteristics of the target person Perform second self attribute learning; The relationship establishment module is used to use the feature data obtained by the first relationship network learning and the second self-attribute learning as the independent variables of the supervised learning algorithm, and the marked historical fraud insurance personnel as the dependent variables to establish the relationship model; A model training module is used to determine a supervised learning algorithm to be constructed when the output of the relational model reaches a preset accuracy rate. 一種處理設備,包括處理器以及用於儲存處理器可執行指令的記憶體,該處理器執行該指令時實現: 獲取待識別人群的關係關聯資料; 基於該關係關聯資料構建該待識別人群的多度關係網路圖資料以及提取該待識別人群的人員特徵資料; 利用構建的有監督學習算法對該待識別人群的多度關係網路圖資料和該人員特徵資料進行識別,確該待識別人群騙保輸出結果;該有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。A processing device includes a processor and a memory for storing processor-executable instructions. When the processor executes the instructions, the processor implements: Obtain relationship data of the population to be identified; Constructing a multi-degree relationship network diagram data of the crowd to be identified based on the relationship association data and extracting personal characteristic data of the crowd to be identified; The constructed supervised learning algorithm is used to identify the multi-relationship network graph data of the crowd to be identified and the characteristics of the person to confirm the fraudulent output of the crowd; the supervised learning algorithm includes the selected target population The multi-relationship network data and personnel characteristics data, the data relationship model obtained by training historical fraud insurance personnel as sample data. 一種伺服器,包括至少一個處理器以及用於儲存處理器可執行指令的記憶體,該處理器執行該指令時實現: 獲取待識別人群的關係關聯資料; 基於該關係關聯資料構建該待識別人群的多度關係網路圖資料以及提取該待識別人群的人員特徵資料; 利用構建的有監督學習算法對該待識別人群的多度關係網路圖資料和該人員特徵資料進行識別,確該待識別人群騙保輸出結果;該有監督學習算法包括採用以選取的目標人群的多度關係網路資料和人員特徵資料、打標的歷史騙保人員作為樣本資料進行訓練得到的資料關係模型。A server includes at least one processor and a memory for storing processor-executable instructions. When the processor executes the instructions, the server implements: Obtain relationship data of the population to be identified; Constructing a multi-degree relationship network diagram data of the crowd to be identified based on the relationship association data and extracting personal characteristic data of the crowd to be identified; The constructed supervised learning algorithm is used to identify the multi-relationship network graph data of the crowd to be identified and the characteristics of the person to confirm the fraudulent output of the crowd; the supervised learning algorithm includes the selected target population The multi-relationship network data and personnel characteristics data, the data relationship model obtained by training historical fraud insurance personnel as sample data.
TW108104900A 2018-04-12 2019-02-14 Data processing method, device, equipment and server for insurance fraud identification TWI686760B (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
CN201810327069.3 2018-04-12
CN201810327069.3A CN108334647A (en) 2018-04-12 2018-04-12 Data processing method, device, equipment and the server of Insurance Fraud identification
??201810327069.3 2018-04-12

Publications (2)

Publication Number Publication Date
TW201944338A true TW201944338A (en) 2019-11-16
TWI686760B TWI686760B (en) 2020-03-01

Family

ID=62934055

Family Applications (1)

Application Number Title Priority Date Filing Date
TW108104900A TWI686760B (en) 2018-04-12 2019-02-14 Data processing method, device, equipment and server for insurance fraud identification

Country Status (3)

Country Link
CN (1) CN108334647A (en)
TW (1) TWI686760B (en)
WO (1) WO2019196552A1 (en)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108334647A (en) * 2018-04-12 2018-07-27 阿里巴巴集团控股有限公司 Data processing method, device, equipment and the server of Insurance Fraud identification
CN109087145A (en) * 2018-08-13 2018-12-25 阿里巴巴集团控股有限公司 Target group's method for digging, device, server and readable storage medium storing program for executing
CN109325525A (en) * 2018-08-31 2019-02-12 阿里巴巴集团控股有限公司 Sample attribute assessment models training method, device and server
CN109447658A (en) * 2018-09-10 2019-03-08 平安科技(深圳)有限公司 The generation of anti-fraud model and application method, device, equipment and storage medium
CN109657890B (en) * 2018-09-14 2023-04-25 蚂蚁金服(杭州)网络技术有限公司 Method and device for determining risk of money transfer fraud
CN109614496B (en) * 2018-09-27 2022-06-17 长威信息科技发展股份有限公司 Low security identification method based on knowledge graph
CN109544379B (en) * 2018-10-30 2024-07-19 平安科技(深圳)有限公司 Unit type determining method and related product
CN109509106B (en) * 2018-10-30 2024-07-19 平安科技(深圳)有限公司 Unit type determining method and related product
CN109801176B (en) * 2019-02-22 2021-04-06 中科软科技股份有限公司 Method, system, electronic device and storage medium for identifying insurance fraud
CN110264371B (en) * 2019-05-10 2024-03-08 创新先进技术有限公司 Information display method, device, computing equipment and computer readable storage medium
CN110428337B (en) * 2019-06-14 2023-01-20 南京极谷人工智能有限公司 Vehicle insurance fraud group partner identification method and device
CN110363406A (en) * 2019-06-27 2019-10-22 上海淇馥信息技术有限公司 Appraisal procedure, device and the electronic equipment of a kind of client intermediary risk
CN110580260B (en) * 2019-08-07 2023-05-26 北京明智和术科技有限公司 Data mining method and device for specific group
CN111415241A (en) * 2020-02-29 2020-07-14 深圳壹账通智能科技有限公司 Method, device, equipment and storage medium for identifying cheater
CN112419074A (en) * 2020-11-13 2021-02-26 中保车服科技服务股份有限公司 Vehicle insurance fraud group identification method and device
CN114170030B (en) * 2021-12-08 2023-09-26 北京百度网讯科技有限公司 Method, apparatus, electronic device and medium for remote damage assessment of vehicle

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7813944B1 (en) * 1999-08-12 2010-10-12 Fair Isaac Corporation Detection of insurance premium fraud or abuse using a predictive software system
JP2003529160A (en) * 2000-03-24 2003-09-30 アクセス ビジネス グループ インターナショナル リミテッド ライアビリティ カンパニー System and method for detecting fraudulent transactions
CN105095238B (en) * 2014-05-04 2019-01-18 ***股份有限公司 For detecting the decision tree generation method of fraudulent trading
US10599999B2 (en) * 2014-06-02 2020-03-24 Yottamine Analytics, Inc. Digital event profile filters based on cost sensitive support vector machine for fraud detection, risk rating or electronic transaction classification
WO2016210122A1 (en) * 2015-06-24 2016-12-29 IGATE Global Solutions Ltd. Insurance fraud detection and prevention system
CN106600413A (en) * 2015-10-19 2017-04-26 阿里巴巴集团控股有限公司 Cheat recognition method and system
CN106600423A (en) * 2016-11-18 2017-04-26 云数信息科技(深圳)有限公司 Machine learning-based car insurance data processing method and device and car insurance fraud identification method and device
CN106803168B (en) * 2016-12-30 2021-04-16 ***股份有限公司 Abnormal transfer detection method and device
CN107066616B (en) * 2017-05-09 2020-12-22 京东数字科技控股有限公司 Account processing method and device and electronic equipment
CN107145587A (en) * 2017-05-11 2017-09-08 成都四方伟业软件股份有限公司 A kind of anti-fake system of medical insurance excavated based on big data
CN107194623B (en) * 2017-07-20 2021-01-05 深圳市分期乐网络科技有限公司 Group partner fraud discovery method and device
CN107785058A (en) * 2017-07-24 2018-03-09 平安科技(深圳)有限公司 Anti- fraud recognition methods, storage medium and the server for carrying safety brain
CN107403326A (en) * 2017-08-14 2017-11-28 云数信息科技(深圳)有限公司 A kind of Insurance Fraud recognition methods and device based on teledata
CN107644098A (en) * 2017-09-29 2018-01-30 马上消费金融股份有限公司 Method, device, equipment and storage medium for identifying fraudulent behavior
CN107730262B (en) * 2017-10-23 2021-09-24 创新先进技术有限公司 Fraud identification method and device
CN107819747B (en) * 2017-10-26 2020-09-18 上海欣方智能***有限公司 Telecommunication fraud association analysis system and method based on communication event sequence
CN108334647A (en) * 2018-04-12 2018-07-27 阿里巴巴集团控股有限公司 Data processing method, device, equipment and the server of Insurance Fraud identification

Also Published As

Publication number Publication date
TWI686760B (en) 2020-03-01
WO2019196552A1 (en) 2019-10-17
CN108334647A (en) 2018-07-27

Similar Documents

Publication Publication Date Title
TWI686760B (en) Data processing method, device, equipment and server for insurance fraud identification
TW201944304A (en) Data processing method, apparatus and device for insurance fraud identification, and server
TWI715879B (en) Method, device and equipment for controlling transaction risk based on graph structure model
TWI746814B (en) Computer readable medium, car insurance risk prediction device and server
US20200202428A1 (en) Graphical structure model-based credit risk control
US11568480B2 (en) Artificial intelligence derived anonymous marketplace
Arundina et al. The predictive accuracy of Sukuk ratings; Multinomial Logistic and Neural Network inferences
US11514369B2 (en) Systems and methods for machine learning model interpretation
Forouzanfar et al. Transport energy demand forecast using multi-level genetic programming
Jang et al. Business failure prediction of construction contractors using a LSTM RNN with accounting, construction market, and macroeconomic variables
US8984022B1 (en) Automating growth and evaluation of segmentation trees
CN111210335A (en) User risk identification method and device and electronic equipment
CN112950347B (en) Resource data processing optimization method and device, storage medium and terminal
CN112200402B (en) Risk quantification method, device and equipment based on risk portrait
Gupta An empirical analysis of default risk for listed companies in India: A comparison of two prediction models
US20190303781A1 (en) System and method for implementing a trust discretionary distribution tool
Calabrese Optimal cut-off for rare events and unbalanced misclassification costs
CN113052679A (en) Model training method, prediction method and device based on multi-view learning and electronic equipment
Nobanee et al. Big data and credit risk assessment: a bibliometric review, current streams, and directions for future research
CN116664306A (en) Intelligent recommendation method and device for wind control rules, electronic equipment and medium
CN115018608A (en) Risk prediction method and device and computer equipment
CN113094595A (en) Object recognition method, device, computer system and readable storage medium
CN118260347B (en) Data acquisition and analysis method and system based on artificial intelligence
CN117893214A (en) Big data wind control processing method and system applying AI
Moon et al. Corporate Bankruptcy Prediction Model using Explainable AI-based Feature Selection