CN113688924A - Abnormal order detection method, device, equipment and medium - Google Patents
Abnormal order detection method, device, equipment and medium Download PDFInfo
- Publication number
- CN113688924A CN113688924A CN202111011117.6A CN202111011117A CN113688924A CN 113688924 A CN113688924 A CN 113688924A CN 202111011117 A CN202111011117 A CN 202111011117A CN 113688924 A CN113688924 A CN 113688924A
- Authority
- CN
- China
- Prior art keywords
- order
- real
- abnormal
- time
- features
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 120
- 238000001514 detection method Methods 0.000 title claims abstract description 116
- 238000012549 training Methods 0.000 claims abstract description 51
- 238000013528 artificial neural network Methods 0.000 claims abstract description 43
- 238000005516 engineering process Methods 0.000 claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 17
- 230000006870 function Effects 0.000 claims description 39
- 238000012545 processing Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 11
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 230000002547 anomalous effect Effects 0.000 claims 1
- 238000013473 artificial intelligence Methods 0.000 abstract description 7
- 230000004913 activation Effects 0.000 description 14
- 210000002569 neuron Anatomy 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 238000007726 management method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 239000004973 liquid crystal related substance Substances 0.000 description 4
- 230000008520 organization Effects 0.000 description 4
- 208000022417 sinus histiocytosis with massive lymphadenopathy Diseases 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an artificial intelligence technology, and discloses an abnormal order detection method, which comprises the following steps: the method comprises the steps of obtaining a historical order set, classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set, extracting order features in the classified order set, training a pre-constructed deep neural network according to the order features to obtain an abnormal detection model, obtaining a real-time order by using a real-time streaming technology, determining the real-time features of the real-time order, and predicting the real-time features by using the abnormal detection model to obtain an abnormal detection result of the real-time order. In addition, the invention also relates to a block chain technology, and the abnormal detection result can be stored in a node of the block chain. The invention also provides an abnormal order detection method and device, electronic equipment and a computer readable storage medium. The invention can solve the problem of low abnormal order detection efficiency.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to an abnormal order detection method and device, electronic equipment and a computer readable storage medium.
Background
With the development of internet technology, product transactions in various fields are currently performed by using online and full-flow integrated software systems, for example, business insurance in the insurance field.
The online order issuing can greatly improve the order issuing efficiency, but the order issuing software can have the risk of causing order issuing exception due to malicious attack of hackers or improper operation of developers and salespeople per se. In the prior art, business personnel are widely used to manually detect whether orders are abnormal, but due to the fact that the number of orders is large, a large number of abnormal orders are always overstocked, instantaneity of abnormal order detection is affected, and meanwhile, manual detection is easy to miss detection, so that detection efficiency is low.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for detecting abnormal orders, and mainly aims to solve the problem of low efficiency of detecting abnormal orders.
In order to achieve the above object, the present invention provides an abnormal order detection method, which includes:
acquiring a historical order set, and classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set;
extracting order features in the classified order set according to a pre-constructed feature database;
training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model;
acquiring a real-time order by using a real-time streaming technology, and determining real-time characteristics of the real-time order according to the characteristic database;
and predicting the real-time characteristics of the real-time order by using the anomaly detection model to obtain the anomaly detection result of the real-time order.
Optionally, the classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set includes:
taking an abnormal order in the historical order set as a negative type sample, and adding an abnormal label;
taking the non-abnormal orders in the historical order set as normal samples, and adding normal labels;
and summarizing the positive sample and the negative sample added with the labels to obtain the classified order set.
Optionally, the training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model includes:
randomly selecting a first number of order features from the positive type samples, selecting a second number of order features from the negative type samples, and summarizing the selected order features to form a training sample set with a preset number;
outputting the prediction rate of the training samples in the training sample set by using the deep neural network;
and calculating a loss value of the prediction rate by using a preset loss function, and when the loss value is greater than a preset loss threshold value, returning to the step of randomly selecting a first number of order features from the positive sample and a second number of order features from the negative sample, and determining that training is finished until the loss value is less than or equal to the loss threshold value, so as to obtain the abnormal detection model.
Optionally, the calculating the loss value of the prediction rate by using a preset loss function:
calculating a loss value of the prediction rate using the following cross entropy loss function:
wherein, Loss is the Loss value, N is the sample number of the training sample set, prediIs the predicted value of the ith sample,/iIs the label of the ith sample.
Optionally, the obtaining a real-time order by using a real-time streaming technology includes:
acquiring a real-time order data stream from a pre-constructed message middleware;
segmenting the real-time order data stream by using a preset batch processing interval to obtain a discrete data stream comprising a plurality of segmented data sets;
extracting a segmentation data set in the discrete data stream by using a preset sliding window in a sliding manner to obtain a batch data set;
and taking the order in the batch data set as the real-time order.
Optionally, before training the pre-constructed deep neural network according to the order features, the method further includes:
setting a characteristic input layer according to the characteristic dimension of the order characteristic;
and constructing a full connection layer, a fitting layer and an output layer behind the characteristic input layer to obtain the deep neural network.
Optionally, the extracting the order features in the classified order set according to a pre-constructed feature database includes:
searching the order type in the characteristic database according to the title of each order in the classified order set;
and determining a characteristic label corresponding to the searched order type, and extracting a target text corresponding to the characteristic label from the classified order set as the order characteristic.
In order to solve the above problem, the present invention further provides an abnormal order detection method apparatus, where the apparatus includes:
the order classification module is used for acquiring a historical order set and classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set;
the characteristic extraction module is used for extracting order characteristics in the classified order set according to a pre-constructed characteristic database;
the model training module is used for training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model;
the real-time characteristic acquisition module is used for acquiring a real-time order by utilizing a real-time streaming technology and determining the real-time characteristics of the real-time order according to the characteristic database;
and the order detection module is used for predicting the real-time characteristics of the real-time order by using the abnormity detection model to obtain an abnormity detection result of the real-time order.
In order to solve the above problem, the present invention also provides an electronic device, including:
a memory storing at least one computer program; and
and the processor executes the computer program stored in the memory to realize the abnormal order detection method.
In order to solve the above problem, the present invention further provides a computer-readable storage medium, in which at least one computer program is stored, and the at least one computer program is executed by a processor in an electronic device to implement the abnormal order detection method described above.
According to the invention, through classifying the mass historical order sets, scattered order features in different orders can be extracted, the order features are deeply mined according to the deep neural network and an effective anomaly detection model is learned, so that the efficiency of anomaly order detection is improved. Meanwhile, the order characteristics of each real-time order are dynamically acquired by using a real-time streaming technology, and compared with the situation that a large number of orders are overstocked due to manual detection, the timeliness of order detection can be improved. Therefore, the abnormal order detection method, the abnormal order detection device, the electronic equipment and the computer readable storage medium provided by the invention can solve the problem of low abnormal order detection efficiency.
Drawings
Fig. 1 is a schematic flow chart illustrating an abnormal order detection method according to an embodiment of the present invention;
FIG. 2 is a functional block diagram of an abnormal order detection apparatus according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device for implementing the abnormal order detection method according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The embodiment of the application provides an abnormal order detection method. The execution subject of the abnormal order detection method includes, but is not limited to, at least one of electronic devices that can be configured to execute the method provided by the embodiment of the present application, such as a server, a terminal, and the like. In other words, the abnormal order detection method may be performed by software or hardware installed in the terminal device or the server device, and the software may be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a Network service, cloud communication, a middleware service, a domain name service, a security service, a Content Delivery Network (CDN), a big data and artificial intelligence platform, and the like.
Fig. 1 is a schematic flow chart of an abnormal order detection method according to an embodiment of the present invention. In this embodiment, the abnormal order detection method includes:
s1, obtaining a historical order set, and classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set.
In the embodiment of the present invention, the historical order set may be a transaction order set in different fields, such as a historical automobile commercial insurance order set in the insurance field, a user shopping order in the e-commerce field, and the like.
Specifically, the classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set includes:
taking an abnormal order in the historical order set as a negative type sample, and adding an abnormal label;
taking the non-abnormal orders in the historical order set as normal samples, and adding normal labels;
and summarizing the positive sample and the negative sample added with the labels to obtain the classified order set.
In the embodiment of the present invention, the abnormal order refers to an order that has been marked as abnormal in the historical order set. For example, in the insurance field, abnormal orders in the historical order set include orders caused by malicious hacking, misoperations by sales personnel, and the like.
The normal label of the positive type sample may be 1, and the abnormal label of the negative type sample may be 0.
In an optional embodiment of the invention, the historical order set is divided into two types of orders according to the states of the orders, so that the training and the performance evaluation of the anomaly detection model are facilitated.
And S2, extracting the order features in the classified order set according to the pre-constructed feature database.
In the embodiment of the present invention, the classified order set includes abnormal orders (negative type samples) and non-abnormal orders (positive type samples) of different products, for example, abnormal orders and non-abnormal orders of products such as car insurance orders and business insurance orders.
Specifically, the pre-built feature database includes preset order types and feature tags to be extracted for different order types. The feature tag is a tag that reflects an order attribute. For example, the feature labels set in the feature database to be extracted from the commercial insurance of the motor vehicle include: insurance policy source, customer number, insured number, applicant number, lot number, associated vehicle number, seat number, site manager number, seat branch center, seat business mode, team long number, task group number, task number, whether to transfer introduction, tertiary organization code, secondary organization code, channel source, business source, system source, special terms, insurance amount, insurance fee, order date and time, whether to apply other series of products, whether to apply electronic insurance policy, business line, etc.
Specifically, the extracting the order features in the classified order set according to the pre-constructed feature database includes:
searching the order type in the characteristic database according to the title of each order in the classified order set;
and determining a characteristic label corresponding to the searched order type, and extracting a target text corresponding to the characteristic label from the classified order set as the order characteristic.
In detail, the searching for the order type in the feature database according to the title of each order in the classified order set includes:
performing word segmentation processing on the order types and the titles of the orders in the characteristic database to obtain a type list and a title list;
constructing an encoding dictionary according to the type list and the title list;
performing vector coding on the type list and the title list by using the coding dictionary to obtain a type vector and a title vector;
and calculating the target similarity of the type vector and the title vector by using a preset cosine similarity calculation formula, and determining the order type corresponding to the type vector with the highest target similarity as the searched order type.
In an alternative embodiment of the invention, the similarity is calculated using the following formula:
wherein a is the type vector and b is the header vector.
In the embodiment of the invention, taking the insurance field as an example, because different insurance orders need to extract different order features, and the order features of each order have fixed feature labels, the feature labels needed by different orders can be quickly positioned by matching the labels in the feature database, and the data processing efficiency is improved.
For example, the order is entitled "XXXX motor vehicle insurance," the type of the order is determined to be motor vehicle insurance by matching the feature database, and the target text in the order is extracted as the order feature: and (3) the sources of the insurance policy: XX, customer number: 001, insured number: 002, applicant no: 003, lot number 004, associated vehicle number: 1234, ….
And S3, training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model.
In an optional embodiment of the present invention, the pre-constructed deep neural network may include six layers of deep neural networks and one output layer. The first layer is a feature input layer, and the number of the neurons is the same as the feature dimension in the order feature set; the second layer is a full connection layer with 64 neurons, and a Relu function is adopted as an activation function; the third layer and the fourth layer are fitting layers which are respectively a batch normalization layer and a Dropout layer and are used for preventing the model from being over-fitted, and the fifth layer and the sixth layer are two depth fully-connected layers which are respectively provided with 32 neurons and 16 neurons and both adopt relu functions as activation functions so as to complete further abstract representation of data characteristics; the output layer is a Dense layer, and a sigmod function is adopted as an activation function. The entire deep neural network may calculate the loss value using a cross entropy loss function.
In detail, before the training the pre-constructed deep neural network according to the order features, the method further includes:
setting a characteristic input layer according to the characteristic dimension of the order characteristic;
and constructing a full connection layer, a fitting layer and an output layer behind the characteristic input layer to obtain the deep neural network.
Wherein, the characteristic dimension refers to the quantity of the order characteristics input each time.
In detail, the training of the pre-constructed deep neural network according to the order features to obtain an anomaly detection model includes:
randomly selecting a first number of order features from the positive type samples, selecting a second number of order features from the negative type samples, and summarizing the selected order features to form a training sample set with a preset number;
outputting the prediction rate of the training samples in the training sample set by using the deep neural network;
and calculating a loss value of the prediction rate by using a preset loss function, and when the loss value is greater than a preset loss threshold value, returning to the step of randomly selecting a first number of order features from the positive sample and a second number of order features from the negative sample, and determining that training is finished until the loss value is less than or equal to the loss threshold value, so as to obtain the abnormal detection model.
In an optional embodiment of the present invention, the outputting the prediction rate of the training samples in the training sample set by using the deep neural network includes:
calculating the prediction rate of the training samples in the deep neural network by using the following prediction formula:
predi=Dense(zi(train)θ,activation=’sigmod’)(zi)
therein, prediFor prediction rate, Dense denotes the output layer of the deep neural network, zi(train) is a training sample, activation ═ sigmod' indicates that the activation function of the output layer is a sigmod function, and θ is a model parameter.
In the embodiment of the present invention, the calculating the loss value of the prediction rate by using a preset loss function:
calculating a loss value of the prediction rate using the following cross entropy loss function:
wherein N is the number of samples in the training sample set, prediIs the predicted value of the ith sample,/iIs the label of the ith sample.
In an alternative embodiment of the present invention, the ratio of the first quantity to the second quantity may be 2: 3, for example, the first quantity and the second quantity may be 40 and 60, respectively.
In the embodiment of the invention, the scattered characteristics of mass historical orders are deeply mined by adopting a deep neural network algorithm, and an effective classification model is learned to be used as an abnormal detection model, so that the detection efficiency of abnormal orders can be improved.
And S4, acquiring the real-time order by using a real-time streaming technology, and determining the real-time characteristics of the real-time order according to the characteristic database.
In the embodiment of the present invention, the real-time Streaming technology may be an Apache Spark Streaming technology, and the Apache Spark Streaming technology may perform Streaming processing on a real-time data stream, and has characteristics of scalability, high throughput, fault tolerance, and the like. The Apache Spark Streaming technique performs high-throughput Streaming computations by converting data streams into elastic Distributed Data Sets (RDDs).
In the embodiment of the present invention, the type of the real-time order is determined by using the feature database, and the target text in the real-time order is extracted as the real-time feature according to the determined type of the order, and the step of extracting the feature is described in S2, which is not described herein again.
Specifically, the obtaining a real-time order by using a real-time streaming technology includes:
acquiring a real-time order data stream from a pre-constructed message middleware;
segmenting the real-time order data stream by using a preset batch processing interval to obtain a discrete data stream comprising a plurality of segmented data sets;
extracting a segmentation data set in the discrete data stream by using a preset sliding window in a sliding manner to obtain a batch data set;
and taking the order in the batch data set as the real-time order.
In an alternative embodiment of the present invention, the message middleware may be Kafka message middleware or the like.
For example, the batch processing interval (batch interval) may be 2S, the real-time order data stream is segmented every 2S to obtain a segmented data set (RDD) including 2S order data, the sliding window (window length) may be 10S, and the batch data set includes 5 segmented data sets. The acquisition capability of the real-time order can be improved through the Apache Spark Streaming technology.
S5, predicting the real-time characteristics of the real-time order by using the anomaly detection model to obtain the anomaly detection result of the real-time order.
Specifically, the predicting the real-time characteristics of the real-time order by using the anomaly detection model to obtain the anomaly detection result of the real-time order includes:
outputting the abnormal probability of the real-time features by using the abnormal detection model;
if the abnormal probability is smaller than a preset abnormal threshold value, determining that the abnormal detection result is that the order is normal;
and if the abnormal probability is larger than or equal to the abnormal threshold, determining that the abnormal detection result is abnormal order and triggering an alarm.
In an alternative embodiment of the present invention, the anomaly threshold is calculated, for example, by the following predictive formula:
pred=Dense(zi(new)θ,activation=’sigmod’)(zi)
where the final pred is the probability of anomaly, zi(new) represents real-time characteristics, the abnormal threshold value can be 0.5, when the output abnormal probability is greater than 0.5, the real-time order is determined to be abnormal, the abnormal order alarming function is triggered, the order information is sent to a manual sales seat, and further the manual sales seat judges and carries out subsequent transaction steps.
The invention is based on the real-time computing technology, receives the basic information of each order to be committed, uses the deep neural network model to automatically and intelligently judge whether an abnormal order exists, further triggers an alarm system if the abnormal order exists, and takes over the subsequent transaction flow by manual sales, thereby greatly improving the efficiency of detecting the abnormal order.
According to the invention, through classifying the mass historical order sets, scattered order features in different orders can be extracted, the order features are deeply mined according to the deep neural network and an effective anomaly detection model is learned, so that the efficiency of anomaly order detection is improved. Meanwhile, the order characteristics of each real-time order are dynamically acquired by using a real-time streaming technology, and compared with the situation that a large number of orders are overstocked due to manual detection, the timeliness of order detection can be improved. Therefore, the abnormal order detection method provided by the invention can solve the problem of low abnormal order detection efficiency.
Fig. 2 is a functional block diagram of an abnormal order detection apparatus according to an embodiment of the present invention.
The abnormal order detection apparatus 100 according to the present invention may be installed in an electronic device. According to the implemented functions, the abnormal order detection apparatus 100 may include an order classification module 101, a feature extraction module 102, a model training module 103, a real-time feature acquisition module 104, and an order detection module 105. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
In the present embodiment, the functions regarding the respective modules/units are as follows:
the order classification module 101 is configured to obtain a historical order set, and classify the historical order set according to a state of each order in the historical order set to obtain a classified order set;
the feature extraction module 102 is configured to extract order features in the classified order set according to a pre-constructed feature database;
the model training module 103 is configured to train a pre-constructed deep neural network according to the order features to obtain an anomaly detection model;
the real-time characteristic obtaining module 104 is configured to obtain a real-time order by using a real-time streaming technology, and determine a real-time characteristic of the real-time order according to the characteristic database;
the order detection module 105 is configured to predict real-time characteristics of the real-time order by using the anomaly detection model, so as to obtain an anomaly detection result of the real-time order.
In detail, the specific implementation of each module of the abnormal order detection apparatus 100 is as follows:
step one, acquiring a historical order set, and classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set.
In the embodiment of the present invention, the historical order set may be a transaction order set in different fields, such as a historical automobile commercial insurance order set in the insurance field, a user shopping order in the e-commerce field, and the like.
Specifically, the classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set includes:
taking an abnormal order in the historical order set as a negative type sample, and adding an abnormal label;
taking the non-abnormal orders in the historical order set as normal samples, and adding normal labels;
and summarizing the positive sample and the negative sample added with the labels to obtain the classified order set.
In the embodiment of the present invention, the abnormal order refers to an order that has been marked as abnormal in the historical order set. For example, in the insurance field, abnormal orders in the historical order set include orders caused by malicious hacking, misoperations by sales personnel, and the like.
The normal label of the positive type sample may be 1, and the abnormal label of the negative type sample may be 0.
In an optional embodiment of the invention, the historical order set is divided into two types of orders according to the states of the orders, so that the training and the performance evaluation of the anomaly detection model are facilitated.
And step two, extracting the order features in the classified order set according to a pre-constructed feature database.
In the embodiment of the present invention, the classified order set includes abnormal orders (negative type samples) and non-abnormal orders (positive type samples) of different products, for example, abnormal orders and non-abnormal orders of products such as car insurance orders and business insurance orders.
Specifically, the pre-built feature database includes preset order types and feature tags to be extracted for different order types. The feature tag is a tag that reflects an order attribute. For example, the feature labels set in the feature database to be extracted from the commercial insurance of the motor vehicle include: insurance policy source, customer number, insured number, applicant number, lot number, associated vehicle number, seat number, site manager number, seat branch center, seat business mode, team long number, task group number, task number, whether to transfer introduction, tertiary organization code, secondary organization code, channel source, business source, system source, special terms, insurance amount, insurance fee, order date and time, whether to apply other series of products, whether to apply electronic insurance policy, business line, etc.
Specifically, the extracting the order features in the classified order set according to the pre-constructed feature database includes:
searching the order type in the characteristic database according to the title of each order in the classified order set;
and determining a characteristic label corresponding to the searched order type, and extracting a target text corresponding to the characteristic label from the classified order set as the order characteristic.
In detail, the searching for the order type in the feature database according to the title of each order in the classified order set includes:
performing word segmentation processing on the order types and the titles of the orders in the characteristic database to obtain a type list and a title list;
constructing an encoding dictionary according to the type list and the title list;
performing vector coding on the type list and the title list by using the coding dictionary to obtain a type vector and a title vector;
and calculating the target similarity of the type vector and the title vector by using a preset cosine similarity calculation formula, and determining the order type corresponding to the type vector with the highest target similarity as the searched order type.
In an alternative embodiment of the invention, the similarity is calculated using the following formula:
wherein a is the type vector and b is the header vector.
In the embodiment of the invention, taking the insurance field as an example, because different insurance orders need to extract different order features, and the order features of each order have fixed feature labels, the feature labels needed by different orders can be quickly positioned by matching the labels in the feature database, and the data processing efficiency is improved.
For example, the order is entitled "XXXX motor vehicle insurance," the type of the order is determined to be motor vehicle insurance by matching the feature database, and the target text in the order is extracted as the order feature: and (3) the sources of the insurance policy: XX, customer number: 001, insured number: 002, applicant no: 003, lot number 004, associated vehicle number: 1234, ….
And step three, training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model.
In an optional embodiment of the present invention, the pre-constructed deep neural network may include six layers of deep neural networks and one output layer. The first layer is a feature input layer, and the number of the neurons is the same as the feature dimension in the order feature set; the second layer is a full connection layer with 64 neurons, and a Relu function is adopted as an activation function; the third layer and the fourth layer are fitting layers which are respectively a batch normalization layer and a Dropout layer and are used for preventing the model from being over-fitted, and the fifth layer and the sixth layer are two depth fully-connected layers which are respectively provided with 32 neurons and 16 neurons and both adopt relu functions as activation functions so as to complete further abstract representation of data characteristics; the output layer is a Dense layer, and a sigmod function is adopted as an activation function. The entire deep neural network may calculate the loss value using a cross entropy loss function.
In detail, before the training the pre-constructed deep neural network according to the order features, the method further includes:
setting a characteristic input layer according to the characteristic dimension of the order characteristic;
and constructing a full connection layer, a fitting layer and an output layer behind the characteristic input layer to obtain the deep neural network.
Wherein, the characteristic dimension refers to the quantity of the order characteristics input each time.
In detail, the training of the pre-constructed deep neural network according to the order features to obtain an anomaly detection model includes:
randomly selecting a first number of order features from the positive type samples, selecting a second number of order features from the negative type samples, and summarizing the selected order features to form a training sample set with a preset number;
outputting the prediction rate of the training samples in the training sample set by using the deep neural network;
and calculating a loss value of the prediction rate by using a preset loss function, and when the loss value is greater than a preset loss threshold value, returning to the step of randomly selecting a first number of order features from the positive sample and a second number of order features from the negative sample, and determining that training is finished until the loss value is less than or equal to the loss threshold value, so as to obtain the abnormal detection model.
In an optional embodiment of the present invention, the outputting the prediction rate of the training samples in the training sample set by using the deep neural network includes:
calculating the prediction rate of the training samples in the deep neural network by using the following prediction formula:
predi=Dense(zi(train)θ,activation=’sigmod’)(zi)
therein, prediFor prediction rate, Dense denotes the output layer of the deep neural network, zi(train) is a training sample, activation ═ sigmod' indicates that the activation function of the output layer is a sigmod function, and θ is a model parameter.
In the embodiment of the present invention, the calculating the loss value of the prediction rate by using a preset loss function:
calculating a loss value of the prediction rate using the following cross entropy loss function:
wherein N is the number of samples in the training sample set, prediIs the predicted value of the ith sample,/iIs the label of the ith sample.
In an alternative embodiment of the present invention, the ratio of the first quantity to the second quantity may be 2: 3, for example, the first quantity and the second quantity may be 40 and 60, respectively.
In the embodiment of the invention, the scattered characteristics of mass historical orders are deeply mined by adopting a deep neural network algorithm, and an effective classification model is learned to be used as an abnormal detection model, so that the detection efficiency of abnormal orders can be improved.
And step four, acquiring a real-time order by using a real-time streaming technology, and determining the real-time characteristics of the real-time order according to the characteristic database.
In the embodiment of the present invention, the real-time Streaming technology may be an Apache Spark Streaming technology, and the Apache Spark Streaming technology may perform Streaming processing on a real-time data stream, and has characteristics of scalability, high throughput, fault tolerance, and the like. The Apache Spark Streaming technique performs high-throughput Streaming computations by converting data streams into elastic Distributed Data Sets (RDDs).
In the embodiment of the present invention, the type of the real-time order is determined by using the feature database, and the target text in the real-time order is extracted as the real-time feature according to the determined type of the order, and the operation of extracting the feature is described in step two, which is not described herein again.
Specifically, the obtaining a real-time order by using a real-time streaming technology includes:
acquiring a real-time order data stream from a pre-constructed message middleware;
segmenting the real-time order data stream by using a preset batch processing interval to obtain a discrete data stream comprising a plurality of segmented data sets;
extracting a segmentation data set in the discrete data stream by using a preset sliding window in a sliding manner to obtain a batch data set;
and taking the order in the batch data set as the real-time order.
In an alternative embodiment of the present invention, the message middleware may be Kafka message middleware or the like.
For example, the batch processing interval (batch interval) may be 2S, the real-time order data stream is segmented every 2S to obtain a segmented data set (RDD) including 2S order data, the sliding window (window length) may be 10S, and the batch data set includes 5 segmented data sets. The acquisition capability of the real-time order can be improved through the Apache Spark Streaming technology.
And fifthly, predicting the real-time characteristics of the real-time order by using the anomaly detection model to obtain the anomaly detection result of the real-time order.
Specifically, the predicting the real-time characteristics of the real-time order by using the anomaly detection model to obtain the anomaly detection result of the real-time order includes:
outputting the abnormal probability of the real-time features by using the abnormal detection model;
if the abnormal probability is smaller than a preset abnormal threshold value, determining that the abnormal detection result is that the order is normal;
and if the abnormal probability is larger than or equal to the abnormal threshold, determining that the abnormal detection result is abnormal order and triggering an alarm.
In an alternative embodiment of the present invention, the anomaly threshold is calculated, for example, by the following predictive formula:
pred=Dense(zi(new)θ,activation=’sigmod’)(zi)
where the final pred is the probability of anomaly, zi(new) represents real-time characteristics, the abnormal threshold value can be 0.5, when the output abnormal probability is greater than 0.5, the real-time order is determined to be abnormal, the abnormal order alarming function is triggered, the order information is sent to a manual sales seat, and further the manual sales seat judges and carries out subsequent transaction steps.
The invention is based on the real-time computing technology, receives the basic information of each order to be committed, uses the deep neural network model to automatically and intelligently judge whether an abnormal order exists, further triggers an alarm system if the abnormal order exists, and takes over the subsequent transaction flow by manual sales, thereby greatly improving the efficiency of detecting the abnormal order.
According to the invention, through classifying the mass historical order sets, scattered order features in different orders can be extracted, the order features are deeply mined according to the deep neural network and an effective anomaly detection model is learned, so that the efficiency of anomaly order detection is improved. Meanwhile, the order characteristics of each real-time order are dynamically acquired by using a real-time streaming technology, and compared with the situation that a large number of orders are overstocked due to manual detection, the timeliness of order detection can be improved. Therefore, the abnormal order detection device provided by the invention can solve the problem of low abnormal order detection efficiency.
Fig. 3 is a schematic structural diagram of an electronic device implementing an abnormal order detection method according to an embodiment of the present invention.
The electronic device may include a processor 10, a memory 11, a communication interface 12 and a bus 13, and may further include a computer program, such as an abnormal order detection program, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, which includes flash memory, removable hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, for example a removable hard disk of the electronic device. The memory 11 may also be an external storage device of the electronic device in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used not only to store application software installed in the electronic device and various types of data, such as codes of an abnormal order detection program, but also to temporarily store data that has been output or is to be output.
The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the electronic device by using various interfaces and lines, and executes various functions and processes data of the electronic device by running or executing programs or modules (e.g., an abnormal order detection program, etc.) stored in the memory 11 and calling data stored in the memory 11.
The communication interface 12 is used for communication between the electronic device and other devices, and includes a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), which are typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable, among other things, for displaying information processed in the electronic device and for displaying a visualized user interface.
The bus 13 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus 13 may be divided into an address bus, a data bus, a control bus, etc. The bus 13 is arranged to enable connection communication between the memory 11 and at least one processor 10 or the like.
Fig. 3 shows only an electronic device having components, and those skilled in the art will appreciate that the structure shown in fig. 3 does not constitute a limitation of the electronic device, and may include fewer or more components than those shown, or some components may be combined, or a different arrangement of components.
For example, although not shown, the electronic device may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management and the like are realized through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
Further, the electronic device may further include a network interface, and optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a bluetooth interface, etc.), which are generally used to establish a communication connection between the electronic device and other electronic devices.
Optionally, the electronic device may further comprise a user interface, which may be a Display (Display), an input unit (such as a Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable, among other things, for displaying information processed in the electronic device and for displaying a visualized user interface.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
The exception order detection program stored in the memory 11 of the electronic device is a combination of instructions, which when executed in the processor 10, may implement:
acquiring a historical order set, and classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set;
extracting order features in the classified order set according to a pre-constructed feature database;
training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model;
acquiring a real-time order by using a real-time streaming technology, and determining real-time characteristics of the real-time order according to the characteristic database;
and predicting the real-time characteristics of the real-time order by using the anomaly detection model to obtain the anomaly detection result of the real-time order.
Specifically, the specific implementation method of the processor 10 for the instruction may refer to the description of the relevant steps in the embodiment corresponding to fig. 1, which is not described herein again.
Further, the electronic device integrated module/unit, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in a computer readable storage medium. The computer readable storage medium may be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying said computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM).
The present invention also provides a computer-readable storage medium, storing a computer program which, when executed by a processor of an electronic device, may implement:
acquiring a historical order set, and classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set;
extracting order features in the classified order set according to a pre-constructed feature database;
training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model;
acquiring a real-time order by using a real-time streaming technology, and determining real-time characteristics of the real-time order according to the characteristic database;
and predicting the real-time characteristics of the real-time order by using the anomaly detection model to obtain the anomaly detection result of the real-time order.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the system claims may also be implemented by one unit or means in software or hardware. The terms second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (10)
1. An abnormal order detection method, characterized in that the method comprises:
acquiring a historical order set, and classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set;
extracting order features in the classified order set according to a pre-constructed feature database;
training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model;
acquiring a real-time order by using a real-time streaming technology, and determining real-time characteristics of the real-time order according to the characteristic database;
and predicting the real-time characteristics of the real-time order by using the anomaly detection model to obtain the anomaly detection result of the real-time order.
2. The abnormal order detection method of claim 1, wherein said classifying the historical order set according to the status of each order in the historical order set to obtain a classified order set comprises:
taking an abnormal order in the historical order set as a negative type sample, and adding an abnormal label;
taking the non-abnormal orders in the historical order set as normal samples, and adding normal labels;
and summarizing the positive sample and the negative sample added with the labels to obtain the classified order set.
3. The abnormal order detection method of claim 2, wherein the training of the pre-constructed deep neural network according to the order features to obtain the abnormal detection model comprises:
randomly selecting a first number of order features from the positive type samples, selecting a second number of order features from the negative type samples, and summarizing the selected order features to form a training sample set with a preset number;
outputting the prediction rate of the training samples in the training sample set by using the deep neural network;
and calculating a loss value of the prediction rate by using a preset loss function, and when the loss value is greater than a preset loss threshold value, returning to the step of randomly selecting a first number of order features from the positive sample and a second number of order features from the negative sample, and determining that training is finished until the loss value is less than or equal to the loss threshold value, so as to obtain the abnormal detection model.
4. The abnormal order detection method of claim 3, wherein the calculating the loss value of the prediction rate using a preset loss function:
calculating a loss value of the prediction rate using the following cross entropy loss function:
wherein, Loss is the Loss value, N is the sample number of the training sample set, prediIs the predicted value of the ith sample,/iIs the label of the ith sample.
5. The abnormal order detection method of claim 1, wherein said obtaining real-time orders using real-time streaming technology comprises:
acquiring a real-time order data stream from a pre-constructed message middleware;
segmenting the real-time order data stream by using a preset batch processing interval to obtain a discrete data stream comprising a plurality of segmented data sets;
extracting a segmentation data set in the discrete data stream by using a preset sliding window in a sliding manner to obtain a batch data set;
and taking the order in the batch data set as the real-time order.
6. The abnormal order detection method of claim 1, wherein prior to training the pre-built deep neural network according to the order characteristics, the method further comprises:
setting a characteristic input layer according to the characteristic dimension of the order characteristic;
and constructing a full connection layer, a fitting layer and an output layer behind the characteristic input layer to obtain the deep neural network.
7. The abnormal order detection method of claim 1, wherein said extracting the order features in the sorted order set according to the pre-constructed feature database comprises:
searching the order type in the characteristic database according to the title of each order in the classified order set;
and determining a characteristic label corresponding to the searched order type, and extracting a target text corresponding to the characteristic label from the classified order set as the order characteristic.
8. An abnormal order detection apparatus, characterized in that the apparatus comprises:
the order classification module is used for acquiring a historical order set and classifying the historical order set according to the state of each order in the historical order set to obtain a classified order set;
the characteristic extraction module is used for extracting order characteristics in the classified order set according to a pre-constructed characteristic database;
the model training module is used for training a pre-constructed deep neural network according to the order features to obtain an anomaly detection model;
the real-time characteristic acquisition module is used for acquiring a real-time order by utilizing a real-time streaming technology and determining the real-time characteristics of the real-time order according to the characteristic database;
and the order detection module is used for predicting the real-time characteristics of the real-time order by using the abnormity detection model to obtain an abnormity detection result of the real-time order.
9. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the anomalous order detection method as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the abnormal order detection method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111011117.6A CN113688924B (en) | 2021-08-31 | 2021-08-31 | Abnormal order detection method, device, equipment and medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111011117.6A CN113688924B (en) | 2021-08-31 | 2021-08-31 | Abnormal order detection method, device, equipment and medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113688924A true CN113688924A (en) | 2021-11-23 |
CN113688924B CN113688924B (en) | 2024-05-31 |
Family
ID=78584339
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111011117.6A Active CN113688924B (en) | 2021-08-31 | 2021-08-31 | Abnormal order detection method, device, equipment and medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113688924B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110874778A (en) * | 2018-08-31 | 2020-03-10 | 阿里巴巴集团控股有限公司 | Abnormal order detection method and device |
CN112465819A (en) * | 2020-12-18 | 2021-03-09 | 平安科技(深圳)有限公司 | Image abnormal area detection method and device, electronic equipment and storage medium |
CN112532643A (en) * | 2020-12-07 | 2021-03-19 | 长春工程学院 | Deep learning-based traffic anomaly detection method, system, terminal and medium |
CN112686762A (en) * | 2020-12-25 | 2021-04-20 | 中国平安人寿保险股份有限公司 | Policy data violation detection method and device, computer equipment and storage medium |
-
2021
- 2021-08-31 CN CN202111011117.6A patent/CN113688924B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110874778A (en) * | 2018-08-31 | 2020-03-10 | 阿里巴巴集团控股有限公司 | Abnormal order detection method and device |
CN112532643A (en) * | 2020-12-07 | 2021-03-19 | 长春工程学院 | Deep learning-based traffic anomaly detection method, system, terminal and medium |
CN112465819A (en) * | 2020-12-18 | 2021-03-09 | 平安科技(深圳)有限公司 | Image abnormal area detection method and device, electronic equipment and storage medium |
CN112686762A (en) * | 2020-12-25 | 2021-04-20 | 中国平安人寿保险股份有限公司 | Policy data violation detection method and device, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN113688924B (en) | 2024-05-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113822494A (en) | Risk prediction method, device, equipment and storage medium | |
CN112801718B (en) | User behavior prediction method, device, equipment and medium | |
CN113592019B (en) | Fault detection method, device, equipment and medium based on multi-model fusion | |
CN115146865A (en) | Task optimization method based on artificial intelligence and related equipment | |
CN111666415A (en) | Topic clustering method and device, electronic equipment and storage medium | |
CN113435202A (en) | Product recommendation method and device based on user portrait, electronic equipment and medium | |
CN111950621A (en) | Target data detection method, device, equipment and medium based on artificial intelligence | |
CN114781832A (en) | Course recommendation method and device, electronic equipment and storage medium | |
CN115081538A (en) | Customer relationship identification method, device, equipment and medium based on machine learning | |
CN114638547A (en) | Enterprise strategy intelligent early warning method and device, electronic equipment and storage medium | |
CN112396547A (en) | Course recommendation method, device, equipment and medium based on unsupervised learning | |
CN113256191A (en) | Classification tree-based risk prediction method, device, equipment and medium | |
CN114612194A (en) | Product recommendation method and device, electronic equipment and storage medium | |
CN113706291A (en) | Fraud risk prediction method, device, equipment and storage medium | |
CN111985545A (en) | Target data detection method, device, equipment and medium based on artificial intelligence | |
CN114281991A (en) | Text classification method and device, electronic equipment and storage medium | |
CN113591881A (en) | Intention recognition method and device based on model fusion, electronic equipment and medium | |
CN113256181A (en) | Risk factor prediction method, device, equipment and medium | |
CN112651782A (en) | Behavior prediction method, device, equipment and medium based on zoom dot product attention | |
CN111950707A (en) | Behavior prediction method, apparatus, device and medium based on behavior co-occurrence network | |
CN112115890A (en) | Drunk driving identification method, device, equipment and medium based on artificial intelligence | |
CN114625340B (en) | Commercial software research and development method, device, equipment and medium based on demand analysis | |
CN114708073B (en) | Intelligent detection method and device for surrounding mark and serial mark, electronic equipment and storage medium | |
CN114996386A (en) | Business role identification method, device, equipment and storage medium | |
CN114722146A (en) | Supply chain asset checking method, device, equipment and medium based on artificial intelligence |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |