TWM590719U - Intelligent conversation management system based on natural language processing - Google Patents
Intelligent conversation management system based on natural language processing Download PDFInfo
- Publication number
- TWM590719U TWM590719U TW108214308U TW108214308U TWM590719U TW M590719 U TWM590719 U TW M590719U TW 108214308 U TW108214308 U TW 108214308U TW 108214308 U TW108214308 U TW 108214308U TW M590719 U TWM590719 U TW M590719U
- Authority
- TW
- Taiwan
- Prior art keywords
- dialogue
- engine
- candidate
- engines
- text content
- Prior art date
Links
Images
Landscapes
- Machine Translation (AREA)
Abstract
一種基於自然語言處理的智能對話管理系統中,管理伺服器在收到對話文本內容和有關於通道伺服器與客戶的參考資料後,根據該參考資料及預定路由規則從N(N≧2)個對話引擎選出N1(N≧N1≧1)個第一候選者;當N1≠1時根據該對話文本內容所含的一個或多個對話特徵從該N1個第一候選者選出N2(N1≧N2≧1)個第二候選者,當N2≠1時根據該N2個第二候選者所對應的對話狀態從其決定出N3(N2≧N3≧1)個第三候選者;將該對話文本內容傳送至第一/二/三候選者;並當N3≠1時從來自該N3個第三候選者的所有回覆文本內容選出最適切者回傳至該通道伺服器。In an intelligent dialogue management system based on natural language processing, the management server receives the dialogue text content and the reference information about the channel server and the client, according to the reference material and the predetermined routing rules from N (N≧2) The dialogue engine selects N1 (N≧N1≧1) first candidates; when N1≠1, N2 (N1≧N2) is selected from the N1 first candidates according to one or more dialogue features contained in the dialogue text content ≧1) second candidates, when N2≠1, N3 (N2≧N3≧1) third candidates are determined from the dialogue state corresponding to the N2 second candidates; the dialogue text content Send to the first/two/three candidates; and when N3≠1, select the most suitable candidate from all the reply text content from the N3 third candidates and send it back to the channel server.
Description
本新型是有關於自然語言的人機對話,特別是指一種基於自然語言處理的智能對話管理系統。The new type relates to man-machine dialogue about natural language, especially refers to an intelligent dialogue management system based on natural language processing.
隨著金融科技的廣泛應用與發展,各家金控企業已逐漸發展出利用人工智慧(AI)的對話機器人以取代線上客服人員。目前,對於可提供各種不同類型之金融服務的企業機構而言,所有業務部門或單位正各自積極發展出與其業務相關聯的對話機器人。在此情況下,若無法有效整合或串聯用於不同金融業務的對話機器人,此等對話機器人不僅無法有效處理或回覆客戶同時詢問有關於多個不同金融業務的對話文本,而且由於各自建立所需的語料庫致使相同的語料資訊並未共享,因而導致的語料資源的浪費。With the widespread application and development of financial technology, various financial control companies have gradually developed conversational robots that use artificial intelligence (AI) to replace online customer service personnel. At present, for enterprise organizations that can provide various types of financial services, all business departments or units are actively developing dialogue robots associated with their businesses. In this case, if the dialogue robots used for different financial businesses cannot be effectively integrated or connected in series, not only can these dialogue robots not effectively process or reply to customers' simultaneous inquiries about the dialogue texts about multiple different financial businesses, but also because they need to establish their own The corpus results in the same corpus information not being shared, which leads to the waste of corpus resources.
因此,如何能以充分利用語料資源的方式提供最適切對話回覆的對話管理實屬當前重要研發課題之一,亦成為目前相關領域極需改進的目標。Therefore, how to provide the most appropriate dialogue response in the way of making full use of corpus resources is one of the current important research and development topics, and it has also become a goal that needs to be improved in related fields.
因此,本新型的一目的,即在提供一種基於自然語言處理的智能對話管理系統,其能克服現有技術的至少一缺點。Therefore, an object of the present invention is to provide an intelligent dialogue management system based on natural language processing, which can overcome at least one shortcoming of the prior art.
於是,本新型所提供的一種基於自然語言處理的智能對話管理系統用於管理與多個不同業務領域有關的人機對話,並包含一語料庫、N(N≧2)個對話引擎、及一管理伺服器。該語料庫儲存有與該等業務領域有關的語料資訊。每一對話引擎連接該語料庫並操作來根據該語料庫所儲存的該語料資訊且利用自然語言處理方式處理與該等業務領域其中一對應者有關之對話。該管理伺服器連接該語料庫及該N個對話引擎,並包括一儲存單元、及一連接該儲存單元的處理單元。該儲存單元儲存N筆分別對應於該N個對話引擎的引擎資訊,每筆引擎資訊包含該N個對話引擎其中一對應者的對話屬性資料、對話狀態資料、及評分權重。該處理單元包括一路由模組、一對話特徵辨識模組、及一連接該路由模組和該對話特徵辨識模組的分派模組。Therefore, an intelligent dialogue management system based on natural language processing provided by the new model is used to manage man-machine dialogues related to multiple different business fields, and includes a corpus, N (N≧2) dialogue engines, and a management server. The corpus stores corpus information related to these business areas. Each dialogue engine connects to the corpus and operates to process dialogues related to one of the counterparts in these business areas using natural language processing based on the corpus information stored in the corpus. The management server is connected to the corpus and the N dialogue engines, and includes a storage unit and a processing unit connected to the storage unit. The storage unit stores N pieces of engine information corresponding to the N dialogue engines, and each piece of engine information includes dialogue attribute data, dialogue state data, and scoring weight of one of the N dialogue engines. The processing unit includes a routing module, a dialog feature recognition module, and a dispatch module connecting the routing module and the dialog feature recognition module.
當該管理伺服器接收到來自一通道伺服器且包含對應於一客戶之對話的對話文本內容以及與該通道伺服器和該客戶相關聯的參考資料的對話資訊時,該處理單元執行以下操作:該路由模組根據該參考資料且利用一預定路由規則,從該N個對話引擎選出N1(N≧N1≧1)個第一候選對話引擎,並將指示出該N1個第一候選對話引擎的路由結果傳送至該分派模組;當N1=1時,該分派模組將該對話文本內容傳送至作為目標對話引擎的該第一候選對話引擎,而當N1≠1時,該分派模組將該路由結果傳送至該對話特徵辨識模組,以致該對話特徵辨識模組利用一經由機器學習方式訓練出的對話特徵模型,辨識出該對話文本內容所含的一個或多個對話特徵,且根據該(等)對話特徵和分別對應於該N1個第一候選對話引擎的N1筆引擎資訊所含的所有對話屬性資料,從該N1個第一候選對話引擎選出N2(N1≧N2≧1)個第二候選對話引擎,並將指示出該N2個第二候選對話引擎的選擇結果傳送至該分派模組;當N2=1時,該分派模組將該對話文本內容傳送至作為目標對話引擎的該第二候選對話引擎,而當N2≠1時,該分派模組根據分別對應於該N2個第二候選對話引擎的N2筆引擎資訊所含的所有對話狀態資料,從該N2個第二候選對話引擎決定出N3(N2≧N3≧1)個第三候選對話引擎,並將該對話文本內容傳送至該N3個第三候選對話引擎,其中當N3=1時,該第三候選對話引擎作為目標對話引擎;及當該管理伺服器接收到來自該目標對話引擎(N1=1或N2=1或N3=1)的回覆文本內容時,該管理伺服器將該回覆文本內容作為目標回覆文本內容傳送至該通道伺服器,而當N3≠1且該管理伺服器接收到來自該N3個第三候選對話引擎其中每一者且包含回覆文本內容和信心度的回覆資料時,該分派模組根據一預定信心度門檻、接收到的所有信心度、及該儲存單元儲存的分別對應於該N3個第三候選對話引擎的N3筆該引擎資訊所含的所有評分權重,從接收到的所有回覆文本內容選出其中一者作為目標回覆文本內容,並將該目標回覆文本內容傳送至該通道伺服器。When the management server receives dialogue information from a channel server and contains dialogue text content corresponding to a client's dialogue and reference data associated with the channel server and the client, the processing unit performs the following operations: The routing module selects N1 (N≧N1≧1) first candidate conversation engines from the N conversation engines based on the reference data and using a predetermined routing rule, and indicates the N1 first candidate conversation engines The routing result is sent to the dispatch module; when N1=1, the dispatch module sends the dialogue text content to the first candidate dialogue engine as the target dialogue engine, and when N1≠1, the dispatch module will The routing result is sent to the dialogue feature recognition module, so that the dialogue feature recognition module uses a dialogue feature model trained by machine learning to recognize one or more dialogue features contained in the dialogue text content, and according to The dialogue feature(s) and all dialogue attribute data contained in the N1 pen engine information corresponding to the N1 first candidate dialogue engines respectively, N2 (N1≧N2≧1) are selected from the N1 first candidate dialogue engines The second candidate dialogue engine, and transmits the selection results indicating the N2 second candidate dialogue engines to the dispatch module; when N2=1, the dispatch module sends the dialogue text content to the target dialogue engine The second candidate dialogue engine, and when N2≠1, the dispatch module selects from the N2 second candidates according to all the dialogue state data contained in the N2 pen engine information corresponding to the N2 second candidate dialogue engines respectively The dialogue engine determines N3 (N2≧N3≧1) third candidate dialogue engines, and transmits the text content of the dialogue to the N3 third candidate dialogue engines. When N3=1, the third candidate dialogue engine serves as Target dialogue engine; and when the management server receives the reply text content from the target dialogue engine (N1=1 or N2=1 or N3=1), the management server uses the reply text content as the target reply text content Sent to the channel server, and when N3≠1 and the management server receives reply data from each of the N3 third candidate dialogue engines and contains reply text content and confidence, the dispatch module is based on A predetermined confidence threshold, all received confidence levels, and all N3 pens of the engine information stored in the storage unit corresponding to the N3 third candidate dialogue engines The content selects one of them as the target reply text content, and sends the target reply text content to the channel server.
本新型之功效在於:以相對較低的系統開發成本,有效達成不同業務領域之語料資訊的共享,以及用於該等業務領域之對話引擎的結合;特別是,對於含有至少一個意圖的對話,藉由路由規則、意圖辨識和智能分派的機制從該等對話引擎決定出目標對話引擎或多個第三候選對話引擎,並藉由限定和擇優機制從來自於該等第三候選對話引擎的所有回覆文本內容選出最適切的目標回覆文本內容提供給客戶。The effect of this new model is to effectively achieve the sharing of corpus information in different business areas and the combination of dialogue engines used in these business areas at a relatively low system development cost; in particular, for dialogues containing at least one intention , By means of routing rules, intent recognition, and intelligent dispatch mechanisms, the target dialogue engine or multiple third candidate dialogue engines are determined from these dialogue engines, and from the third candidate dialogue engines through the restriction and preference mechanism All reply text content selects the most appropriate target reply text content to provide to customers.
在本新型被詳細描述之前,應當注意在以下的說明內容中,類似的元件是以相同的編號來表示。Before the present invention is described in detail, it should be noted that in the following description, similar elements are denoted by the same number.
參閱圖1,所繪示的本新型實施例基於自然語言處理的智能對話管理系統100用於管理與多個不同業務領域有關的人機對話。舉例來說,在本實施例中,該等業務領域可包含由例如一金控機構所提供的一般客服業務(不執行任何任務型操作)、行動銀行業務(可執行如轉帳、換匯等的任務型操作)、基金/***推薦業務(可執行如基金申購或***申請等的任務型操作)、催繳業務(可提供如主動提醒客戶繳款、紀錄繳款時間等的操作)等,但不以此例為限。該智能對話管理系統100可被實施成一電腦系統,且例如包含一語料庫21、一對話紀錄資料庫22、N(例如,N=4,但不在此限)個對話引擎11-14、及一管理伺服器3。Referring to FIG. 1, the illustrated intelligent
在本實施例中,該語料庫21儲存有與該等業務領域有關的語料資訊。更具體地,該語料資訊例如可包含與該等業務領域有關的意圖語料及實體(Entity)辭庫(例如包含人名、地名、機構名、日期時間、專有名詞等),但不以此例為限。In this embodiment, the
該對話紀錄資料庫22儲存有與該N個對話引擎有關的所有歷史對話紀錄資料。更具體地,對於每次歷史對話,對應的歷史對話紀錄資料例如可包含代表發動該次歷史對話的通道伺服器的通道識別碼、代表對應客戶的客戶識別碼、該次歷史對話的發生時間、該對應客戶於該次歷史對話的對話文本內容、對應(目標)對話引擎11/12/13/14於該次歷史對話的(目標)回覆文本內容、指示出該對應(目標)對話引擎11/12/13/14的對話狀態的對話狀態資料(即,對話尚未結束或對話已結束)。The
在本實施例中,該N(N=4)個對話引擎11-14連接該與料庫21,並分別操作來根據該語料庫21所儲存的該語料資訊(即,該意圖語料與該實體詞庫)且利用自然語言處理方式處理與該等業務領域有關之對話。更具體地,例如,該對話引擎11係配置來處理有關於一般客服業務的對話;該對話引擎12係配置來處理有關於行動銀行業務的對話;該對話引擎13係配置來處理有關於基金/***推薦業務的對話;及該對話引擎14係配置來處理有關於催收業務的對話,但不以此例為限。In this embodiment, the N (N=4) dialogue engines 11-14 are connected to the
附帶一提的是,該語料庫21與該對話紀錄資料庫22可進一步整合成一大數據平台(圖未示),此大數據平台可進一步蒐集該等對話引擎11-14在處理對話後所獲得的意圖詞彙與實體詞彙,以及分析該對話紀錄資料庫22所儲存的所有資料內容,以便根據蒐集到的意圖詞彙與實體詞彙以及分析結果,更新該語料庫21所儲存的該意圖語料和該實體辭庫。Incidentally, the
該管理伺服器3連接該語料庫21、該對話紀錄資料庫11和該等對話引擎11-14,並且包括一儲存單元31及一處理單元32。在本實施例中,如圖2所示,該儲存單元31儲存了N(例如,N=4)筆分別對應於該等對話引擎11-14的引擎資訊。每筆引擎資料例如包含對應的對話引擎11/12/13/14的對話屬性資料、對話狀態資料(其指示出該對應的對話引擎11/12/13/14所處理的最近歷史對話已結束或尚未結束)及評分權重。舉例來說,該對話引擎11的對話屬性資料例如指示出一般(但不涉及任何任務型操作)詢問與意圖;該對話引擎12的對話屬性資料例如指示與行動銀行業務有關的詢問、意圖及任務型操作;該對話引擎13的對話屬性資料例如指示語基金/***推薦業務的有關的詢問、意圖及任務型操作;及該對話引擎14的對話屬性資料例如指示與催收業務有關的提醒、詢問及意圖。如圖3所示,該處理單元32例如包含一路由模組321、一對話特徵辨識模組322、及一連接該路由模組321和該對話特徵辨識模組322的分派模組323。該路由模組321、該對話特徵辨識模組322及該分派模組323其中每一者可被實施成軟體、韌體及硬體其中任一組合的形式。The management server 3 is connected to the
以下,將參閱圖1至圖5來示例地說明當該管理伺服器3接收到來自一通道伺服器200的對話資訊時,該智能對話管理系統100如何執行一智能對話管理程序。在本實施例中,該通道伺服器200例如為一網銀伺服器或一行動銀行伺服器,但不以此例為限。該對話資訊例如包含該通道伺服器200的通道識別碼、對應於一客戶之對話的對話文本內容、及與該通道伺服器200和該客戶相關聯的參考資料。更具體地,該參考資料例如包含該通道伺服器200的通道識別碼、該客戶的客戶識別碼、及代表該客戶的屬性或消費習慣的客戶標籤,但不以此例為限。該智能對話管理程序包含以下步驟S41~S53。Hereinafter, referring to FIGS. 1 to 5, an example will be described to explain how the intelligent
首先,在步驟S41中,該路由模組321根據該參考資料且利用一預定路由規則,從該N(N=4)個對話引擎11-14選出N1(4≧N1≧1)個第一候選對話引擎,並將指示出該N1個第一候選引擎的路由結果傳送至該分派模組323。在本實施例中,該預定路由規則與通道識別碼及客戶標籤其中至少一者相關聯。舉例來說,根據該預定路由規則所規範:當該通道識別碼代表一網銀伺服器時,處理一般客服業務之對話的該對話引擎11(作為單一(N1=1)個第一候選對話引擎)會被選出;當該通道識別碼代表一行動銀行伺服器時,該對話引擎11及處理行動銀行服務之對話的該對話引擎12(作為兩(N1=2)個第一候選對話引擎)會被選出;及當該通道識別碼代表一行動銀行伺服器且同時該客戶標籤指示出該客戶為男性之單身貴族時,該對話引擎11、該對話引擎12和處理基金/***推薦業務之對話的該對話引擎13(作為三(N1=3)個第一候選對話引擎)會被選出。First, in step S41, the
然後,在步驟S42中,該分派模組323確認來自該路由模組321的該路由結果指示的N1是否等於1。若確認結果為N1=1時,該第一候選對話引擎作為目標對話引擎,且流程進行至步驟S43,否則(即,N1≠1),流程進行至步驟S44。Then, in step S42, the
在步驟S43中,該分派模組323將該對話文本內容傳送至該目標對話引擎。舉例來說,在上述範例中,若該通道識別碼代表一網銀伺服器時,該路由模組321在步驟S41中僅會選出該對話引擎11,於是該分派模組323會將該對話文本內容傳送至作為該目標對話引擎的該對話引擎11。In step S43, the
在步驟S44中,該分派模組323將該路由結果傳送至該對話特徵辨識模組322。In step S44, the
接著,在步驟S45中,該對話特徵辨識模組322利用一經由機器學習方式訓練出的對話特徵模型,辨識出該對話文本內容所含的一個或多個對話特徵。在本實施例中,該對話特徵辨識模組辨識出的每一對話特徵屬於情緒特徵、關鍵字特徵、實體特徵及意圖特徵其中一者。Next, in step S45, the dialogue
然後,在步驟S46中,該對話特徵辨識模組322根據該(等)對話特徵和該儲存單元儲存的分別對應於該N1個第一候選對話引擎的N1筆引擎資訊所含的所有對話屬性資料,從該N1個第一候選對話引擎選出N2(N1≧N2≧1)個第二候選對話引擎,並將指示出該N2個第二候選對話引擎的選擇結果傳送至該分派模組323。更具體地,該對話特徵辨識模組322藉由判定每一個第一候選對話引擎的該對話屬性資料是否匹配於該(等)對話特徵來決定出該N2個第二候選對話引擎。Then, in step S46, the dialog
舉例來說,在上述範例中,若該通道識別碼代表一行動銀行伺服器時以致該路由模組321在步驟S41中會選出該等對話引擎11,12,例如,該對話文本內容例如為「我昨天轉帳幾次」的情況下,該對話特徵辨識模組322在步驟S45中會辨識出該對話文本內容的該等對話特徵例如包含屬於實體特徵的「我」和「昨天」、屬於關鍵字特徵的「轉帳」、及屬於意圖特徵的「(詢問)幾次」,於是,該對話特徵辨識模組322會在步驟S46中僅判定出該對話引擎12的該對話屬性資料匹配上述該等對話特徵,但以此例為限。For example, in the above example, if the channel identification code represents a mobile banking server so that the
之後,在步驟S47中,該分派模組323確認來自該對話特徵識別模組322的該選擇結果指示的N2是否等於1。若確認結果為N2=1時,該第二候選對話引擎作為目標對話引擎,且流程進行至步驟S43,否則(即,N2≠1),流程進行至步驟S48。Then, in step S47, the
在步驟S48中,該分派模組323根據該儲存單元31儲存的分別對應於該N2個第二候選對話引擎的N2筆引擎資訊所含的所有對話狀態資料,從該N2個第二候選對話引擎決定出N3(N2≧N3≧1)個第三候選對話引擎,並將該對話文本內容傳送至該N3個第三候選對話引擎。更具體地,該分派模組323藉由確定該儲存單元儲存的對應於每一個第二候選對話引擎的該對話狀態資料是否指示出最近歷史對話尚未結束來決定出該N3個第三候選對話引擎。換言之,與指示出最近歷史對話尚未結束之該對話狀態資料對應的該第二候選對話引擎會優先被選為第三候選對話引擎。值得注意的是,當N3=1時,該第三候選對話引擎作為目標對話引擎。In step S48, the
跟隨在步驟S43或步驟S48之後的步驟S49中,該目標對話引擎(即,N1=1時的該第一候選對話引擎,N2=1時的該第二候選對話引擎或N3=1時的該第三候選對話引擎),或者N3≠1時的每一個第三候選對話引擎根據來自該管理伺服器3的該對話文本內容、及該語料庫21所儲存的該語料資訊且利用自然語言處理方式,產生回覆文本內容,且以現有信心度評估方式獲得該回覆文本內容的信心度,並將含有該回覆文本內容合該信心度的回覆資料傳送至該管理伺服器3。Following in step S49 following step S43 or step S48, the target dialogue engine (ie, the first candidate dialogue engine when N1=1, the second candidate dialogue engine when N2=1 or the second candidate dialogue engine when N3=1 Third candidate dialogue engine), or each third candidate dialogue engine when N3≠1 according to the content of the dialogue text from the management server 3 and the corpus information stored in the
之後,該分派模組323確認是否僅接收到(來自該目標對話引擎的)單一的回覆文本內容(步驟S50)。若該確認結果為否定時(即,該分派模組323接收到分別來自該N3(N2≧N3≧1)個第三候選對話引擎的N3筆回覆資料時),流程進行至步驟S51。相反地若該確認結果為肯定時,在此情況下,該分派模組323會將接收到的該單一回覆文本內容作為目標回覆文本內容,則流程進行步驟S52。After that, the
在步驟S51中,該分派模組323根據一預定信心度門檻(例如,0.8,但不以此例為限)、接收到的所有信心度、及該儲存單元31儲存的分別對應於該N3個第三候選對話引擎的N3筆該引擎資訊所含的所有評分權重,從接收到的所有回覆文本內容選出其中一者作為目標回覆文本內容。更具體地,該分派模組323首先,對於每一個第三候選對話引擎,計算出接收自該第三候選對話引擎的該信心度與該儲存單元31儲存的對應於該第三候選對話引擎的該評分權重的乘積,且使該乘積代表該第三候選對話引擎所具有的評分值;然後從該N3個第三候選對話引擎選出其中一個具有最大評分值且來自其的信心度高於該預定信心度門檻的第三候選對話引擎作為目標對話引擎;最後將來自該目標對話引擎的該回覆資料所含的該回覆文本內容確認為該目標回覆文本內容,藉此,不僅可過濾掉信心度太低(低於該預定信心度門檻)的回覆文本內容,而且選出的目標回覆文本內容是最適切於該對話文本內容(具有最大評分值)。In step S51, the
跟隨在步驟S51之後的步驟S52中,該分派模組323將該目標回覆文本內容傳送至該通道伺服器200。Following step S51 following step S51, the
最後,在步驟S53中,該分派模組323將與該客戶、該目標對話引擎和該通道伺服器200有關的該對話文本內容及該目標回覆文本內容新增地記錄於該對話紀錄資料庫22,並根據該對話文本內容與該目標回覆文本內容決定是否更新該儲存單元31儲存的對應於該目標對話引擎的該對話狀態資料。舉例來說,若該對話文本內容例如為「我要轉帳給我媽媽」且目標回覆文本內容例如為「請問要轉多少錢」時,該分派模組323會將該儲存單元31儲存的對應於該目標對話引擎的該引擎資訊所含的對話狀態資料維持在“指示出最近歷史對話尚未結束”。在另一範例中,若該對話文本內容例如為「是的」且目標回覆文本內容例如為「已為您轉帳給媽媽5000元」時,該分派模組323會將該儲存單元31儲存的對應於該目標對話引擎的該引擎資訊所含的對話狀態資料從“指示出最近歷史對話尚未結束”更新為“指示出最近歷史對話已結束”。Finally, in step S53, the
至此,該智能對話管理程序執行完成。附帶一提的是,該客戶後續可進一步根據該次對話所接獲的該目標回覆文本內容給予滿意度評分,此滿意度評分可經由該通道伺服器200提供給該管理伺服器3。於是,該管理伺服器3可在每一預定期間(例如,每兩週,但不以此例為限)蒐集有關該目標對話引擎的所有滿意度評分作為進一步調整該儲存單元31儲存的對應於該目標對話引擎的評分權重的依據。At this point, the execution of the intelligent dialogue management program is completed. Incidentally, the customer can further give a satisfaction score based on the target reply text content received in the conversation, and the satisfaction score can be provided to the management server 3 via the
綜上所述,該智能對話管理系統100能以相對較低的系統開發成本,有效達成不同業務領域之語料資訊的共享,以及用於該等業務領域之對話引擎的結合;特別是,對於含有至少一個意圖的對話,藉由路由規則、意圖辨識和智能分派的機制從該等對話引擎決定出目標對話引擎或多個第三候選對話引擎,並藉由限定高於該預定信心度門檻和選擇具有最大評分值的機制從來自於該等第三候選對話引擎的所有回覆文本內容選出最適切的目標回覆文本內容提供給客戶。故確實能達成本新型的目的。In summary, the intelligent
惟以上所述者,僅為本新型之實施例而已,當不能以此限定本新型實施之範圍,凡是依本新型申請專利範圍及專利說明書內容所作之簡單的等效變化與修飾,皆仍屬本新型專利涵蓋之範圍內。However, the above are only examples of the new model. When the scope of the new model cannot be limited by this, any simple equivalent changes and modifications made according to the patent application scope and patent specification content of the new model are still regarded as Within the scope of this new patent.
100‧‧‧智能對話管理系統
11-14‧‧‧對話引擎
21‧‧‧語料庫
22‧‧‧度話紀錄資料庫
3‧‧‧管理伺服器
31‧‧‧儲存單元
32‧‧‧處理單元
321‧‧‧路由模組
322‧‧‧對話特徵識別模組
323‧‧‧分派模組
200‧‧‧通道伺服器
S41-S53‧‧‧步驟
100‧‧‧Intelligent dialogue management system
11-14‧‧‧
本新型之其他的特徵及功效,將於參照圖式的實施方式中清楚地呈現,其中: 圖1是一方塊圖,示例性地繪示本新型實施例的智能對話管理系統; 圖2是一示意圖,示例性地說明該實施例的一管理伺服器的一儲存單元儲存的資料內容; 圖3是一方塊圖,示例性地說明該實施例的該管理伺服器的一處理單元的配置;及 圖4及圖5是流程圖,示例性地說明該實施例如何對於接收到的對話資訊執行一智能對話管理程序。 Other features and functions of the present invention will be clearly presented in the embodiments with reference to the drawings, in which: FIG. 1 is a block diagram exemplarily illustrating an intelligent dialogue management system of an embodiment of the present invention; 2 is a schematic diagram exemplarily illustrating the content of data stored in a storage unit of a management server of this embodiment; 3 is a block diagram exemplarily illustrating the configuration of a processing unit of the management server of this embodiment; and FIG. 4 and FIG. 5 are flowcharts exemplarily illustrating how this embodiment executes an intelligent dialog management procedure for the received dialog information.
100‧‧‧智能對話管理系統 100‧‧‧Intelligent dialogue management system
11-14‧‧‧對話引擎 11-14‧‧‧ Dialogue Engine
21‧‧‧語料庫 21‧‧‧ Corpus
22‧‧‧度話紀錄資料庫 22‧‧‧ Degree Record Database
3‧‧‧管理伺服器 3‧‧‧ Management Server
31‧‧‧儲存單元 31‧‧‧Storage unit
32‧‧‧處理單元 32‧‧‧Processing unit
200‧‧‧通道伺服器 200‧‧‧channel server
Claims (6)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108214308U TWM590719U (en) | 2019-10-30 | 2019-10-30 | Intelligent conversation management system based on natural language processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108214308U TWM590719U (en) | 2019-10-30 | 2019-10-30 | Intelligent conversation management system based on natural language processing |
Publications (1)
Publication Number | Publication Date |
---|---|
TWM590719U true TWM590719U (en) | 2020-02-11 |
Family
ID=70414699
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW108214308U TWM590719U (en) | 2019-10-30 | 2019-10-30 | Intelligent conversation management system based on natural language processing |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWM590719U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI752367B (en) * | 2019-10-30 | 2022-01-11 | 中國信託商業銀行股份有限公司 | Intelligent dialogue management method and system based on natural language processing |
-
2019
- 2019-10-30 TW TW108214308U patent/TWM590719U/en unknown
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI752367B (en) * | 2019-10-30 | 2022-01-11 | 中國信託商業銀行股份有限公司 | Intelligent dialogue management method and system based on natural language processing |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11416867B2 (en) | Machine learning system for transaction reconciliation | |
US20210019357A1 (en) | Analytics-driven recommendation engine | |
US11017038B2 (en) | Identification and evaluation white space target entity for transaction operations | |
US9575936B2 (en) | Word cloud display | |
WO2021150591A1 (en) | System, apparatus and methods for providing an intent suggestion to a user in a text-based conversational experience with user feedback | |
US10783877B2 (en) | Word clustering and categorization | |
US20210374749A1 (en) | User profiling based on transaction data associated with a user | |
US9367526B1 (en) | Word classing for language modeling | |
CN112799747A (en) | Intelligent assistant evaluation and recommendation method, system, terminal and readable storage medium | |
US10805465B1 (en) | Predictive customer service support system and method | |
US10922633B2 (en) | Utilizing econometric and machine learning models to maximize total returns for an entity | |
US20230274282A1 (en) | Transaction tracking and fraud detection using voice and/or video data | |
CN115129832A (en) | Facilitating troubleshooting conversations over product support issues via a chat robot | |
TWI752367B (en) | Intelligent dialogue management method and system based on natural language processing | |
KR20220083183A (en) | System for supporting bidding strategy establishment using personalization based business recommendation algorithm and method thereof | |
US20230334339A1 (en) | Supervised and/or unsupervised machine learning models for supplementing records stored in a database for retrieval | |
Oraby et al. | Modeling and computational characterization of twitter customer service conversations | |
US11550786B1 (en) | System, method, and computer program for converting a natural language query to a structured database update statement | |
US20230351121A1 (en) | Method and system for generating conversation flows | |
TWM590719U (en) | Intelligent conversation management system based on natural language processing | |
US20230222527A1 (en) | Omnichannel data processing and analysis | |
CN112650829A (en) | Customer service processing method and device | |
CN115982391A (en) | Information processing method and device | |
US11860824B2 (en) | Graphical user interface for display of real-time feedback data changes | |
US11966570B2 (en) | Automated processing and dynamic filtering of content for display |