CN113221984A - Method, device and equipment for analyzing and predicting drunk driving behaviors of user and storage medium - Google Patents

Method, device and equipment for analyzing and predicting drunk driving behaviors of user and storage medium Download PDF

Info

Publication number
CN113221984A
CN113221984A CN202110474075.3A CN202110474075A CN113221984A CN 113221984 A CN113221984 A CN 113221984A CN 202110474075 A CN202110474075 A CN 202110474075A CN 113221984 A CN113221984 A CN 113221984A
Authority
CN
China
Prior art keywords
drunk driving
user
data
preset
track
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.)
Pending
Application number
CN202110474075.3A
Other languages
Chinese (zh)
Inventor
张霖
朱磊
俞丽娟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Priority to CN202110474075.3A priority Critical patent/CN113221984A/en
Publication of CN113221984A publication Critical patent/CN113221984A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/21—Design, administration or maintenance of databases
    • G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering 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
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Evolutionary Computation (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Marketing (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Databases & Information Systems (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Evolutionary Biology (AREA)
  • Operations Research (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Development Economics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Game Theory and Decision Science (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Educational Administration (AREA)
  • General Health & Medical Sciences (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a method, a device, equipment and a storage medium for analyzing and predicting drunk driving behaviors of a user, which are used for improving the recognition efficiency of drunk driving behavior risks of the user. The method for analyzing and predicting the drunk driving behavior of the user comprises the following steps: cleaning the position-based service track data of the user and the action hotspot track data of the user according to a preset data cleaning rule to obtain cleaned position-based service track data and cleaned action hotspot track data; determining a risk characteristic factor data set based on case work order information, cleaned location-based service track data and cleaned action hotspot track data; carrying out drunk driving analysis and prediction on the risk characteristic factor data set through a preset drunk driving prediction model to obtain a drunk driving prediction result; and generating and pushing the drunk driving evaluation information of the user according to the drunk driving prediction result. In addition, the invention also relates to a block chain technology, and the drunk driving evaluation information of the user can be stored into the block chain link points.

Description

Method, device and equipment for analyzing and predicting drunk driving behaviors of user and storage medium
Technical Field
The invention relates to the field of artificial intelligence neural networks, in particular to a method, a device, equipment and a storage medium for analyzing and predicting drunk driving behaviors of a user.
Background
The drunk driving risk identification refers to that when a traffic accident happens, people systematically and continuously know the drunk driving risk and analyze the potential reasons of the traffic risk accident. The drunk driving risk identification process comprises two links of sensing risk and analyzing risk.
Conventional risk identification schemes use models that generally form specific rules for determining drunk driving behavior based on historical empirical data. That is, the data available to detect user drunk driving behavior is often limited to historically collected claims data and small amounts of external data. In a vehicle loss compensation business, drunk driving groups often have some specific behavior patterns, and through a traditional risk identification scheme, the accuracy rate of predicting the drunk driving behavior risk of a user is low, and the vehicle loss compensation risk caused by drunk driving is high.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for analyzing and predicting drunk driving behaviors of a user, which are used for improving the recognition accuracy and recognition efficiency of drunk driving behavior risks of the user and reducing vehicle compensation loss risks.
In order to achieve the above object, a first aspect of the present invention provides a method for analyzing and predicting drunk driving behaviors of a user, including: receiving a drunk driving behavior analysis request of a user, and acquiring user detail data according to the drunk driving behavior analysis request of the user, wherein the user detail data comprises case work order information, position-based service track data of the user and action hotspot track data of the user; performing data cleaning on the position-based service track data of the user and the action hotspot track data of the user according to a preset data cleaning rule to obtain cleaned position-based service track data and cleaned action hotspot track data; performing data analysis and statistical processing on the case work order information, the cleaned location-based service track data and the cleaned action hotspot track data to obtain a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors; carrying out drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result; and generating user drunk driving evaluation information according to the drunk driving prediction result, and pushing the user drunk driving evaluation information to a target terminal so that the target terminal processes corresponding vehicle compensation business according to the user drunk driving evaluation information.
Optionally, in a first implementation manner of the first aspect of the present invention, the performing data cleaning on the location-based service trajectory data of the user and the action hotspot trajectory data of the user according to a preset data cleaning rule to obtain cleaned location-based service trajectory data and cleaned action hotspot trajectory data includes: performing data analysis on a preset data cleaning rule according to a preset data format to obtain an analyzed data cleaning rule, wherein the analyzed data cleaning rule comprises a track data type, a plurality of track fields and a data cleaning conversion expression corresponding to each track field; and respectively carrying out data deduplication, abnormal data deletion, null value data cleaning and data format standardization processing on the position-based service track data of the user and the action hotspot track data of the user in sequence based on the track data type, the plurality of track fields and the data cleaning conversion expression corresponding to each track field to obtain cleaned position-based service track data and cleaned action hotspot track data.
Optionally, in a second implementation manner of the first aspect of the present invention, the performing data analysis and statistical processing on the case work order information, the cleaned location-based service trajectory data, and the cleaned action hotspot trajectory data to obtain a risk characteristic factor dataset, where the risk characteristic factor dataset is used to indicate a plurality of characteristic factors related to drunk driving behaviors, includes: performing data analysis on the case work order information to obtain the case occurrence time and the case occurrence place; screening the peripheral environment information of the case location according to the case occurrence location and a preset location range, and combining the case occurrence time and the peripheral environment information of the case location into case characteristic factors; respectively reading a plurality of position service interest point data and a plurality of action hotspot interest point data which are associated with a target place from the cleaned position-based service track data and the cleaned action hotspot track data according to a preset feature extraction strategy, wherein the target place comprises an entertainment place and a dining place; respectively carrying out clustering analysis on the plurality of position service interest point data and the plurality of action hotspot interest point data according to preset dimensions to obtain position service characteristic factors and action hotspot characteristic factors; combining the case characteristic factors, the location service characteristic factors and the action hotspot characteristic factors into a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors.
Optionally, in a third implementation manner of the first aspect of the present invention, the analyzing and predicting drunk driving on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result includes: inputting the risk characteristic factor data set into a preset drunk driving prediction model, and performing characteristic coding and vectorization processing on the risk characteristic factor data set through a characteristic extraction layer in the drunk driving prediction model to obtain a plurality of embedded vector characteristics, wherein the embedded vector characteristics comprise a position service characteristic vector set and a motion hotspot characteristic vector set; performing cross feature combination on the plurality of embedded vector features through a factor decomposition machine layer in the drunk driving prediction model to obtain a first output result; performing characteristic full-connection processing on the plurality of embedded vector characteristics through a deep neural network layer in the drunk driving prediction model to obtain a second output result; and carrying out drunk driving analysis and prediction on the first output result and the second output result through an activation function layer in the drunk driving prediction model to obtain a drunk driving prediction result, and mapping and storing the drunk driving prediction result and the user drunk driving behavior analysis request into a preset memory data table.
Optionally, in a fourth implementation manner of the first aspect of the present invention, the generating user drunk driving evaluation information according to the drunk driving prediction result, and pushing the user drunk driving evaluation information to a target terminal, so that the target terminal processes a corresponding vehicle compensation service according to the user drunk driving evaluation information, includes: matching a preset information classification table according to the drunk driving prediction result to obtain an information configuration template, and generating user drunk driving evaluation information based on the information configuration template and the case work order information; the drunk driving evaluation information of the user is pushed to a target terminal through a preset message queue, so that the target terminal processes corresponding vehicle compensation services according to the drunk driving evaluation information of the user, and the vehicle compensation services comprise vehicle damage compensation services and personal injury compensation services.
Optionally, in a fifth implementation manner of the first aspect of the present invention, before the receiving the drunk driving behavior analysis request of the user, and obtaining user detail data according to the drunk driving behavior analysis request of the user, where the user detail data includes case work order information, location-based service trajectory data of the user, and action hotspot trajectory data of the user, the drunk driving behavior analysis and prediction method for the user further includes: extracting position-based service behavior track sample data and action hotspot behavior track sample data from a preset drunk driving refusal claim case and a preset non-drunk driving refusal claim case; converting the position service-based behavior track sample data and the action hotspot behavior track sample data into position service-based sample characteristic factors and action hotspot sample characteristic factors according to the moment dimension and the space dimension; and training an initial deep neural network model according to the position service sample characteristic factors and the action hotspot sample characteristic factors to obtain the preset drunk driving prediction model.
Optionally, in a sixth implementation manner of the first aspect of the present invention, after the generating the user drunk driving evaluation information according to the drunk driving prediction result, and pushing the user drunk driving evaluation information to a target terminal, so that the target terminal processes a corresponding vehicle compensation service according to the user drunk driving evaluation information, the method for analyzing and predicting drunk driving behavior of the user further includes: constructing a user behavior portrait according to the user detail data, the drunk driving prediction result, the user drunk driving evaluation information and the vehicle compensation service; and storing the user behavior portrait in a preset graph database, and sending the user behavior portrait to a target terminal.
The second aspect of the present invention provides a device for analyzing and predicting drunk driving behaviors of a user, including: the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for receiving a drunk driving behavior analysis request of a user and acquiring user detail data according to the drunk driving behavior analysis request of the user, and the user detail data comprises case work order information, position-based service track data of the user and action hotspot track data of the user; the cleaning module is used for carrying out data cleaning on the position-based service track data of the user and the action hotspot track data of the user according to a preset data cleaning rule to obtain cleaned position-based service track data and cleaned action hotspot track data; the analysis module is used for carrying out data analysis and statistical processing on the case work order information, the cleaned location-based service track data and the cleaned action hotspot track data to obtain a risk characteristic factor data set, and the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors; the prediction module is used for carrying out drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result; the generation module is used for generating user drunk driving evaluation information according to the drunk driving prediction result and pushing the user drunk driving evaluation information to a target terminal so that the target terminal processes corresponding vehicle compensation business according to the user drunk driving evaluation information.
Optionally, in a first implementation manner of the second aspect of the present invention, the cleaning module is specifically configured to: performing data analysis on a preset data cleaning rule according to a preset data format to obtain an analyzed data cleaning rule, wherein the analyzed data cleaning rule comprises a track data type, a plurality of track fields and a data cleaning conversion expression corresponding to each track field; and respectively carrying out data deduplication, abnormal data deletion, null value data cleaning and data format standardization processing on the position-based service track data of the user and the action hotspot track data of the user in sequence based on the track data type, the plurality of track fields and the data cleaning conversion expression corresponding to each track field to obtain cleaned position-based service track data and cleaned action hotspot track data.
Optionally, in a second implementation manner of the second aspect of the present invention, the analysis module is specifically configured to: performing data analysis on the case work order information to obtain the case occurrence time and the case occurrence place; screening the peripheral environment information of the case location according to the case occurrence location and a preset location range, and combining the case occurrence time and the peripheral environment information of the case location into case characteristic factors; respectively reading a plurality of position service interest point data and a plurality of action hotspot interest point data which are associated with a target place from the cleaned position-based service track data and the cleaned action hotspot track data according to a preset feature extraction strategy, wherein the target place comprises an entertainment place and a dining place; respectively carrying out clustering analysis on the plurality of position service interest point data and the plurality of action hotspot interest point data according to preset dimensions to obtain position service characteristic factors and action hotspot characteristic factors; combining the case characteristic factors, the location service characteristic factors and the action hotspot characteristic factors into a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors.
Optionally, in a third implementation manner of the second aspect of the present invention, the prediction module is specifically configured to: inputting the risk characteristic factor data set into a preset drunk driving prediction model, and performing characteristic coding and vectorization processing on the risk characteristic factor data set through a characteristic extraction layer in the drunk driving prediction model to obtain a plurality of embedded vector characteristics, wherein the embedded vector characteristics comprise a position service characteristic vector set and a motion hotspot characteristic vector set; performing cross feature combination on the plurality of embedded vector features through a factor decomposition machine layer in the drunk driving prediction model to obtain a first output result; performing characteristic full-connection processing on the plurality of embedded vector characteristics through a deep neural network layer in the drunk driving prediction model to obtain a second output result; and carrying out drunk driving analysis and prediction on the first output result and the second output result through an activation function layer in the drunk driving prediction model to obtain a drunk driving prediction result, and mapping and storing the drunk driving prediction result and the user drunk driving behavior analysis request into a preset memory data table.
Optionally, in a fourth implementation manner of the second aspect of the present invention, the generating module is specifically configured to: matching a preset information classification table according to the drunk driving prediction result to obtain an information configuration template, and generating user drunk driving evaluation information based on the information configuration template and the case work order information; the drunk driving evaluation information of the user is pushed to a target terminal through a preset message queue, so that the target terminal processes corresponding vehicle compensation services according to the drunk driving evaluation information of the user, and the vehicle compensation services comprise vehicle damage compensation services and personal injury compensation services.
Optionally, in a fifth implementation manner of the second aspect of the present invention, the device for analyzing and predicting drunk driving behavior of a user further includes: the extraction module is used for extracting position-based service behavior track sample data and action hotspot behavior track sample data from a preset drunk driving claim rejection case and a preset non-drunk driving claim rejection case; the conversion module is used for converting the position-based service behavior trace sample data and the action hotspot behavior trace sample data into position-based service sample characteristic factors and action hotspot sample characteristic factors according to the moment dimension and the space dimension; and the training module is used for training an initial deep neural network model according to the position service sample characteristic factors and the action hotspot sample characteristic factors to obtain the preset drunk driving prediction model.
Optionally, in a sixth implementation manner of the second aspect of the present invention, the device for analyzing and predicting drunk driving behavior of a user further includes: the construction module is used for constructing a user behavior portrait according to the user detail data, the drunk driving prediction result, the user drunk driving evaluation information and the vehicle compensation business; and the storage module is used for storing the user behavior portrait in a preset map database and sending the user behavior portrait to a target terminal.
The third aspect of the present invention provides a user drunk driving behavior analysis and prediction device, including: a memory and at least one processor, the memory having instructions stored therein; the at least one processor calls the instructions in the memory to enable the user drunk driving behavior analysis and prediction device to execute the user drunk driving behavior analysis and prediction method.
A fourth aspect of the present invention provides a computer-readable storage medium, having stored therein instructions, which, when run on a computer, cause the computer to execute the above-mentioned method for analyzing and predicting drunk driving behavior of a user.
According to the technical scheme provided by the invention, a drunk driving behavior analysis request of a user is received, and user detail data is obtained according to the drunk driving behavior analysis request of the user, wherein the user detail data comprises case work order information, position-based service track data of the user and action hotspot track data of the user; performing data cleaning on the position-based service track data of the user and the action hotspot track data of the user according to a preset data cleaning rule to obtain cleaned position-based service track data and cleaned action hotspot track data; performing data analysis and statistical processing on the case work order information, the cleaned location-based service track data and the cleaned action hotspot track data to obtain a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors; carrying out drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result; and generating user drunk driving evaluation information according to the drunk driving prediction result, and pushing the user drunk driving evaluation information to a target terminal so that the target terminal processes corresponding vehicle compensation business according to the user drunk driving evaluation information. In the embodiment of the invention, the drunk driving related factor characteristics are analyzed on case work order information, the user position-based service track data and the user action hotspot track data, and the drunk driving related factor characteristics are cross-analyzed through the preset abnormality detection algorithm of the drunk driving prediction model, so that the users with high drunk driving risk can be screened out, the recognition accuracy and recognition efficiency of the drunk driving behavior risk of the users are improved, and the vehicle compensation loss risk is reduced.
Drawings
Fig. 1 is a schematic diagram of an embodiment of a method for analyzing and predicting drunk driving behaviors of a user according to an embodiment of the invention;
fig. 2 is a schematic diagram of another embodiment of the method for analyzing and predicting drunk driving behaviors of a user according to the embodiment of the invention;
fig. 3 is a schematic diagram of an embodiment of a drunk driving behavior analysis and prediction device for a user according to an embodiment of the present invention;
fig. 4 is a schematic diagram of another embodiment of the drunk driving behavior analysis and prediction device for the user in the embodiment of the invention;
fig. 5 is a schematic diagram of an embodiment of a user drunk driving behavior analysis and prediction device in the embodiment of the invention.
Detailed Description
The embodiment of the invention provides a method, a device, equipment and a storage medium for analyzing and predicting drunk driving behaviors of a user, which are used for improving the recognition accuracy and recognition efficiency of drunk driving behavior risks of the user and reducing vehicle compensation loss risks.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Furthermore, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
For convenience of understanding, a specific flow of an embodiment of the present invention is described below, and referring to fig. 1, an embodiment of a method for analyzing and predicting drunk driving behaviors of a user in an embodiment of the present invention includes:
101. receiving a drunk driving behavior analysis request of a user, and acquiring user detail data according to the drunk driving behavior analysis request of the user, wherein the user detail data comprises case work order information, position-based service track data of the user and action hotspot track data of the user.
The case work order information may include the case occurrence time, the case occurrence place and the user report telephone number. The user detail data may include case work order information, location-based service trajectory data of the user, and action hotspot trajectory data of the user, and may further include user age, user residence address information, and user drunk driving times, which is not limited herein. Specifically, the server receives a drunk driving behavior analysis request of the user, and the server performs parameter analysis on the drunk driving behavior analysis request of the user to obtain parameter analysis data; the server checks the parameter analysis data to obtain a check result; when the verification result is a preset value, the server determines that the verification result is verification passing, and the server extracts the user identification from the parameter analysis data; the server generates a first query statement according to a structured query language grammar rule, a user identification and a preset to-be-processed work order state; the server executes the first query statement and reads case work order information from a preset work order database; the server generates a second query statement according to the grammar rule of the structured query language, the user report telephone number and the preset track data type; the server executes a second query statement, and reads the position-based service track data of the user and the user action hotspot track data from a preset track database; and the server assembles the case work order information, the position-based service track data of the user and the action hotspot track data of the user into user detail data.
It should be noted that the location-based service trajectory data of the user and the action hotspot trajectory data of the user are both obtained and stored by the server from a preset data source according to a preset data extraction rule. The server can also pre-store the case work order information, the user position-based service track data and the user action hotspot track data in a preset memory database. The server sets the user identification or the user report telephone number as a target key, and queries a preset memory database according to the target key to obtain user detail data, wherein the user detail data comprises case work order information, user position-based service track data and user action hotspot track data.
It is to be understood that the executing subject of the present invention may be a drunk driving behavior analyzing and predicting device, and may also be a terminal or a server, which is not limited herein. The embodiment of the present invention is described by taking a server as an execution subject.
102. And performing data cleaning on the position-based service track data of the user and the action hotspot track data of the user according to a preset data cleaning rule to obtain cleaned position-based service track data and cleaned action hotspot track data.
The preset data cleaning rule comprises a track data type, a plurality of track fields and a data cleaning conversion expression corresponding to each track field. For example, the data type corresponding to the location-based service trajectory data of the user may be type _1, and the data type corresponding to the action hotspot trajectory data of the user may be type _ 2. The user location based service track field has a plurality of track fields and the user action hotspot track data has a plurality of track fields. Specifically, the server respectively performs data deduplication, abnormal data deletion, null value data cleaning and unified data format on the position-based service trajectory data of the user and the action hotspot trajectory data of the user according to a preset data cleaning rule, so as to obtain cleaned position-based service trajectory data and cleaned action hotspot trajectory data.
103. And carrying out data analysis and statistical processing on the case work order information, the cleaned position-based service track data and the cleaned action hotspot track data to obtain a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors.
The risk characteristic factor data set is drunk driving activity factors processed according to case work order information, cleaned position-based service track data and cleaned action hotspot track data, namely a plurality of characteristic factors related to drunk driving behaviors. Specifically, the server performs data analysis and statistical processing on case work order information, cleaned location-based service track data and cleaned action hotspot track data to obtain a characteristic factor related to location-based service (namely, location service characteristic factor), a characteristic factor related to action hotspot (namely, action hotspot characteristic factor) and a characteristic factor related to case (namely, case characteristic factor); the server combines the characteristic factors related to the location-based service, the characteristic factors related to the action hot spots and the characteristic factors related to the case into a risk characteristic factor data set, and the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors. The characteristic factors related to the location-based service comprise the activity range of the user, the activity place of the user, the activity time of the user, the occurrence frequency and the occurrence time of the user in the entertainment place, and the occurrence frequency and the occurrence time of the user in the catering place; the characteristic factors related to the action hot spot include: the connecting time of the user at the action hotspot of the entertainment place, the connecting times and the connecting time of the user at the catering place; case-related characteristic factors: the reporting time, the reporting place and the surrounding environment of the reporting place.
104. And carrying out drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result.
The preset drunk driving prediction model is used for indicating a pre-trained model. For example, the preset drunk driving prediction model may be a random forest classification model, or may also be a bayesian classification model or a logistic regression model, and is not limited herein. Further, the server analyzes and predicts the drunk driving of the risk characteristic factor data set through a trained deep neural network-based FM model deep FM model (namely, a preset drunk driving prediction model), and a drunk driving prediction result is obtained. Specifically, the server performs feature coding and feature vectorization processing on a risk feature factor data set through a preset drunk driving prediction model to obtain a location service feature vector set and a behavior hotspot feature vector set, the server performs feature processing on the location service feature vector set and the behavior hotspot feature vector set through a hidden layer to obtain a location service hidden layer feature vector set and a behavior hotspot feature vector set, and the server performs cross feature combination on the location service feature vector set and the behavior hotspot feature vector set through a factor decomposition machine FM layer to obtain a first output result. The server carries out cross feature combination full-connection processing on the position service hidden layer feature vector set and the action hot spot hidden layer feature vector set through a deep neural network layer to obtain a second output result; and the server analyzes and predicts the drunk driving of the first output result and the second output result through an activation function sigmoid layer of the neural network to obtain a drunk driving prediction result, wherein the value range of the drunk driving prediction result is between 0 and 1.
105. And generating user drunk driving evaluation information according to the drunk driving prediction result, and pushing the user drunk driving evaluation information to the target terminal so that the target terminal processes corresponding vehicle compensation business according to the user drunk driving evaluation information.
And the drunk driving prediction result and the drunk driving evaluation information of the user have a one-to-one correspondence relationship. The vehicle compensation service includes a vehicle damage compensation service and may also include a personal injury compensation service, which is not limited herein. The value range of the drunk driving prediction result can be 0-1. For example, if the drunk driving prediction result is 0, the evaluation information of drunk driving of the user generated by the server is "the user does not have the risk of drunk driving", and if the drunk driving prediction result is 0.9, the evaluation information of drunk driving of the user generated by the server is "the user has the risk of drunk driving". Specifically, the server generates user drunk driving evaluation information according to the drunk driving prediction result, and classifies the user drunk driving evaluation information according to the drunk driving prediction result to obtain different types of messages to be pushed; the server pushes different types of messages to be pushed to the target terminal according to the message types through a preset message queue, so that the target terminal processes corresponding vehicle compensation services according to the drunk driving evaluation information of the user. For example, after the target terminal checks the vehicle compensation service, the target terminal issues a compensation amount corresponding to the vehicle compensation service to the user.
In the embodiment of the invention, the drunk driving related factor characteristics are analyzed on case work order information, the user position-based service track data and the user action hotspot track data, and the drunk driving related factor characteristics are cross-analyzed through the preset abnormality detection algorithm of the drunk driving prediction model, so that the users with high drunk driving risk can be screened out, the recognition accuracy and recognition efficiency of the drunk driving behavior risk of the users are improved, and the vehicle compensation loss risk is reduced.
Referring to fig. 2, another embodiment of the method for analyzing and predicting drunk driving behaviors of a user according to the embodiment of the present invention includes:
201. receiving a drunk driving behavior analysis request of a user, and acquiring user detail data according to the drunk driving behavior analysis request of the user, wherein the user detail data comprises case work order information, position-based service track data of the user and action hotspot track data of the user.
The execution process of step 201 is similar to the execution process of step 101, and detailed description thereof is omitted here.
202. And performing data cleaning on the position-based service track data of the user and the action hotspot track data of the user according to a preset data cleaning rule to obtain cleaned position-based service track data and cleaned action hotspot track data.
Namely, the server carries out standardized processing on the position-based service track data of the user and the action hotspot track data of the user, and the normalization of the data is improved. Optionally, the server performs data analysis on a preset data cleaning rule according to a preset data format to obtain an analyzed data cleaning rule, where the analyzed data cleaning rule includes a track data type, multiple track fields, and a data cleaning conversion expression corresponding to each track field, where the track data type has a one-to-many correspondence with the multiple track fields and the data cleaning conversion expressions corresponding to each track field, and the track fields correspond to the data cleaning conversion expressions one by one. The preset data format may be a JSON format of a JS object numbered notation, or a YAML format of data serialization, and is not limited herein. The preset data cleansing rule may be stored in a preset memory database (e.g., remote service dictionary redis), or may be stored in a preset file.
Then, the server sequentially performs data deduplication, abnormal data deletion, null value data cleaning and data format standardization processing on the position service track data of the user and the action hotspot track data of the user respectively based on the track data type, a plurality of track fields and a data cleaning conversion expression corresponding to each track field to obtain cleaned position service track data and cleaned action hotspot track data. Further, the server fills missing values and cleans unnecessary data for the track data of the user based on the position service and the track data of the action hot spot of the user respectively based on the track data type, a plurality of track fields and a data cleaning conversion expression corresponding to each track field. For example, the server deletes location information greater than a preset distance from the location-based service trajectory data of the user, and further deletes location trajectory data such as hotels, theaters, books, gas stations, etc., having a preset distance of more than 500 meters from the location-based service trajectory data of the user.
203. And carrying out data analysis and statistical processing on the case work order information, the cleaned position-based service track data and the cleaned action hotspot track data to obtain a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors.
It can be understood that each feature factor in the risk feature factor dataset is used for indicating the risk behaviors of the user such as wide social activity range, more times of the user appearing in the dining place, more times of the user appearing in the entertainment place, preference of the user activity at night, deviation of the driving track of the user from the non-main road and the like. Optionally, the server performs data analysis on the case work order information to obtain the case occurrence time and the case occurrence location, for example, the case occurrence time of a traffic accident is 2021-04-1022: 20, and the case occurrence location is a suburb; the server screens the surrounding environment information of the case place according to the case occurrence place and a preset place range, and combines the case occurrence time and the surrounding environment information of the case place into case characteristic factors, for example, whether the surrounding environment information of the case place comprises a bar or an entertainment place or not; the server respectively reads a plurality of position service interest point data and a plurality of action hotspot interest point data which are associated with a target place from cleaned position-based service track data and cleaned action hotspot track data according to a preset feature extraction strategy, wherein the target place comprises an entertainment place and a catering place, and the entertainment place and the catering place are risk places related to drunk driving behaviors; the server performs clustering analysis on the plurality of position service interest point data and the plurality of action hotspot interest point data respectively according to preset dimensions to obtain position service characteristic factors and action hotspot characteristic factors, and further, the server performs clustering analysis on the plurality of position service interest point data and the plurality of action hotspot interest point data respectively according to preset dimensions based on a K-means clustering algorithm, a density-based clustering algorithm or a mean shift clustering algorithm to obtain position service characteristic factors and action hotspot characteristic factors. The preset dimensions comprise a time dimension and a space dimension; the server combines the case characteristic factors, the location service characteristic factors and the action hotspot characteristic factors into a risk characteristic factor data set, and the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors.
204. And carrying out drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result.
Further, the server can convert the drunk driving prediction result into percentage, namely the value range of the drunk driving prediction result is between 0 and 100. And, as the drunk driving prediction result increases, the risk value indicating the drunk driving behavior of the user increases. Optionally, the server inputs the risk characteristic factor data set into a preset drunk driving prediction model, and performs characteristic coding and vectorization processing on the risk characteristic factor data set through a characteristic extraction layer in the drunk driving prediction model to obtain a plurality of embedded vector characteristics, wherein the plurality of embedded vector characteristics comprise a position service characteristic vector set and a motion hotspot characteristic vector set; the server carries out cross feature combination on the embedded vector features through a factor decomposition machine layer in the drunk driving prediction model to obtain a first output result; the server performs characteristic full-connection processing on the plurality of embedded vector characteristics through a deep neural network layer in the drunk driving prediction model to obtain a second output result; and the server analyzes and predicts the drunk driving of the first output result and the second output result through an activation function layer in the drunk driving prediction model to obtain a drunk driving prediction result, and the drunk driving prediction result and the drunk driving behavior analysis request of the user are mapped and stored into a preset memory data table.
The method includes the steps that a server trains a preset drunk driving prediction model in advance, and further extracts position-based service behavior track sample data and action hotspot behavior track sample data from preset drunk driving claim rejection cases and preset non-drunk driving claim rejection cases; the server converts the position-based service behavior track sample data and the action hotspot behavior track sample data into position-based service sample characteristic factors and action hotspot sample characteristic factors according to the moment dimension and the space dimension; and the server trains the initial deep neural network model according to the position service sample characteristic factors and the action hotspot sample characteristic factors to obtain a preset drunk driving prediction model.
205. And generating user drunk driving evaluation information according to the drunk driving prediction result, and pushing the user drunk driving evaluation information to the target terminal so that the target terminal processes corresponding vehicle compensation business according to the user drunk driving evaluation information.
Further, the value range of the drunk driving prediction result may be between 0 and 1, or between 0 and 100, and is not limited herein. For example, if the drunk driving prediction result is 100, the target terminal refuses to process the vehicle damage compensation service and the personal injury compensation service according to the drunk driving evaluation information of the user, and if the drunk driving prediction result is 25, the target terminal confirms and submits the vehicle damage compensation service and the personal injury compensation service according to the drunk driving evaluation information of the user, and the server performs service operations such as auditing, settling and the like on the vehicle damage compensation service and the personal injury compensation service.
Optionally, the server matches a preset information classification table according to the drunk driving prediction result to obtain an information configuration template, and generates user drunk driving evaluation information based on the information configuration template and case work order information; the server pushes the drunk driving evaluation information of the user to the target terminal through a preset message queue, so that the target terminal processes corresponding vehicle compensation services according to the drunk driving evaluation information of the user, and the vehicle compensation services comprise vehicle damage compensation services and personal injury compensation services.
206. And constructing a user behavior portrait according to the user detail data, the drunk driving prediction result, the drunk driving evaluation information of the user and the vehicle compensation service.
Specifically, the server determines a user fact label according to the user detail data and the vehicle compensation service; the server determines a user auxiliary label according to the drunk driving evaluation information and drunk driving prediction results of the user; and the server performs entity analysis and entity relationship analysis processing on the user fact label and the user auxiliary label to obtain the user behavior portrait.
207. And storing the user behavior portrait in a preset graph database, and sending the user behavior portrait to a target terminal.
The preset graph database may be neo4j, or may be a distributed database flockDB, or may be a graphical database allogorrap, or may be other types of graph databases, which is not limited herein. Specifically, the server acquires a user identifier; the server updates the user behavior portrait to a preset graph database based on the user identification; and the server calls a preset data transmission interface to send the user behavior portrait to the target terminal so that the target terminal draws and displays the user behavior portrait.
It can be understood that the server can find important information such as driving behavior patterns and travel laws of the user by analyzing user behavior track information such as position-based service track data of the user and action hotspot track data of the user, and improve the recognition rate of drunk driving behaviors of the user by establishing rich user figures, thereby effectively preventing the risk that the user obtains vehicle damage compensation due to drunk driving, and reducing vehicle compensation loss of target enterprises (such as insurance enterprises).
In the embodiment of the invention, the drunk driving related factor characteristics are analyzed on case work order information, the user position-based service track data and the user action hotspot track data, and the drunk driving related factor characteristics are cross-analyzed through the preset abnormality detection algorithm of the drunk driving prediction model, so that the users with high drunk driving risk can be screened out, the recognition accuracy and recognition efficiency of the drunk driving behavior risk of the users are improved, and the vehicle compensation loss risk is reduced.
The above description of the method for analyzing and predicting drunk driving behavior of the user in the embodiment of the present invention, and the following description of the device for analyzing and predicting drunk driving behavior of the user in the embodiment of the present invention refer to fig. 3, where an embodiment of the device for analyzing and predicting drunk driving behavior of the user in the embodiment of the present invention includes:
the acquisition module 301 is configured to receive a drunk driving behavior analysis request of a user, and acquire user detail data according to the drunk driving behavior analysis request of the user, where the user detail data includes case work order information, location-based service trajectory data of the user, and action hotspot trajectory data of the user;
a cleaning module 302, configured to perform data cleaning on the location-based service trajectory data of the user and the action hotspot trajectory data of the user according to a preset data cleaning rule, to obtain cleaned location-based service trajectory data and cleaned action hotspot trajectory data;
the analysis module 303 is configured to perform data analysis and statistical processing on the case work order information, the cleaned location-based service trajectory data, and the cleaned action hotspot trajectory data to obtain a risk characteristic factor data set, where the risk characteristic factor data set is used to indicate a plurality of characteristic factors related to drunk driving behaviors;
the prediction module 304 is used for performing drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result;
the generating module 305 is configured to generate the user drunk driving evaluation information according to the drunk driving prediction result, and push the user drunk driving evaluation information to the target terminal, so that the target terminal processes the corresponding vehicle compensation service according to the user drunk driving evaluation information.
Further, the driving intention risk assessment report is stored in the blockchain database, which is not limited herein.
In the embodiment of the invention, the drunk driving related factor characteristics are analyzed on case work order information, the user position-based service track data and the user action hotspot track data, and the drunk driving related factor characteristics are cross-analyzed through the preset abnormality detection algorithm of the drunk driving prediction model, so that the users with high drunk driving risk can be screened out, the recognition accuracy and recognition efficiency of the drunk driving behavior risk of the users are improved, and the vehicle compensation loss risk is reduced.
Referring to fig. 4, another embodiment of the device for analyzing and predicting drunk driving behaviors of a user according to the embodiment of the present invention includes:
the acquisition module 301 is configured to receive a drunk driving behavior analysis request of a user, and acquire user detail data according to the drunk driving behavior analysis request of the user, where the user detail data includes case work order information, location-based service trajectory data of the user, and action hotspot trajectory data of the user;
a cleaning module 302, configured to perform data cleaning on the location-based service trajectory data of the user and the action hotspot trajectory data of the user according to a preset data cleaning rule, to obtain cleaned location-based service trajectory data and cleaned action hotspot trajectory data;
the analysis module 303 is configured to perform data analysis and statistical processing on the case work order information, the cleaned location-based service trajectory data, and the cleaned action hotspot trajectory data to obtain a risk characteristic factor data set, where the risk characteristic factor data set is used to indicate a plurality of characteristic factors related to drunk driving behaviors;
the prediction module 304 is used for performing drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result;
the generating module 305 is configured to generate the user drunk driving evaluation information according to the drunk driving prediction result, and push the user drunk driving evaluation information to the target terminal, so that the target terminal processes the corresponding vehicle compensation service according to the user drunk driving evaluation information.
Optionally, the cleaning module 302 may be further specifically configured to:
performing data analysis on a preset data cleaning rule according to a preset data format to obtain an analyzed data cleaning rule, wherein the analyzed data cleaning rule comprises a track data type, a plurality of track fields and a data cleaning conversion expression corresponding to each track field;
based on the track data type, a plurality of track fields and a data cleaning conversion expression corresponding to each track field, data deduplication, abnormal data deletion, null data cleaning and data format standardization processing are sequentially performed on the position service track data of the user and the action hotspot track data of the user respectively, and cleaned position service track data and cleaned action hotspot track data are obtained.
Optionally, the analysis module 303 may be further specifically configured to:
carrying out data analysis on the case work order information to obtain the case occurrence time and the case occurrence place;
screening the peripheral environment information of the case location according to the case occurrence location and a preset location range, and combining the case occurrence time and the peripheral environment information of the case location into case characteristic factors;
respectively reading a plurality of position service interest point data and a plurality of action hotspot interest point data which are associated with a target place from the cleaned position-based service track data and the cleaned action hotspot track data according to a preset feature extraction strategy, wherein the target place comprises an entertainment place and a catering place;
respectively carrying out clustering analysis on the multiple position service interest point data and the multiple action hotspot interest point data according to preset dimensions to obtain position service characteristic factors and action hotspot characteristic factors;
and combining the case characteristic factors, the location service characteristic factors and the action hotspot characteristic factors into a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors.
Optionally, the prediction module 304 may be further specifically configured to:
inputting the risk characteristic factor data set into a preset drunk driving prediction model, and performing characteristic coding and vectorization processing on the risk characteristic factor data set through a characteristic extraction layer in the drunk driving prediction model to obtain a plurality of embedded vector characteristics, wherein the embedded vector characteristics comprise a position service characteristic vector set and a motion hotspot characteristic vector set;
performing cross feature combination on the plurality of embedded vector features through a factor decomposition machine layer in the drunk driving prediction model to obtain a first output result;
performing characteristic full-connection processing on the plurality of embedded vector characteristics through a deep neural network layer in the drunk driving prediction model to obtain a second output result;
and carrying out drunk driving analysis and prediction on the first output result and the second output result through an activation function layer in the drunk driving prediction model to obtain a drunk driving prediction result, and mapping and storing the drunk driving prediction result and the drunk driving behavior analysis request of the user into a preset memory data table.
Optionally, the generating module 305 may be further specifically configured to:
matching a preset information classification table according to the drunk driving prediction result to obtain an information configuration template, and generating user drunk driving evaluation information based on the information configuration template and case work order information;
the drunk driving evaluation information of the user is pushed to the target terminal through a preset message queue, so that the target terminal processes corresponding vehicle compensation services according to the drunk driving evaluation information of the user, and the vehicle compensation services comprise vehicle damage compensation services and personal injury compensation services.
Optionally, the device for analyzing and predicting drunk driving behavior of the user may further include:
the extraction module 306 is configured to extract location-based service behavior trace sample data and action hotspot behavior trace sample data from a preset drunk driving abstaining case and a preset non-drunk driving abstaining case;
a conversion module 307, configured to convert the sample data based on the location service behavior trace and the sample data based on the action hotspot behavior trace into a feature factor based on the location service sample and a feature factor based on the action hotspot sample according to a moment dimension and a space dimension;
the training module 308 is configured to train the initial deep neural network model according to the location service sample feature factors and the action hotspot sample feature factors to obtain a preset drunk driving prediction model.
Optionally, the device for analyzing and predicting drunk driving behavior of the user may further include:
the construction module 309 is used for constructing a user behavior portrait according to the user detail data, the drunk driving prediction result, the drunk driving evaluation information and the vehicle compensation service;
the storage module 310 is configured to store the user behavior representation into a preset graph database, and send the user behavior representation to the target terminal.
In the embodiment of the invention, the drunk driving related factor characteristics are analyzed on case work order information, the user position-based service track data and the user action hotspot track data, and the drunk driving related factor characteristics are cross-analyzed through the preset abnormality detection algorithm of the drunk driving prediction model, so that the users with high drunk driving risk can be screened out, the recognition accuracy and recognition efficiency of the drunk driving behavior risk of the users are improved, and the vehicle compensation loss risk is reduced.
Fig. 3 and 4 describe the user drunk driving behavior analysis and prediction device in the embodiment of the present invention in detail from the perspective of modularization, and the user drunk driving behavior analysis and prediction device in the embodiment of the present invention is described in detail from the perspective of hardware processing.
Fig. 5 is a schematic structural diagram of a user drunk driving behavior analysis and prediction device 500 according to an embodiment of the present invention, which may generate relatively large differences due to different configurations or performances, and may include one or more processors (CPUs) 510 (e.g., one or more processors) and a memory 520, one or more storage media 530 (e.g., one or more mass storage devices) storing applications 533 or data 532. Memory 520 and storage media 530 may be, among other things, transient or persistent storage. The program stored in the storage medium 530 may include one or more modules (not shown), each of which may include a series of instruction operations in the user drunk driving behavior analysis prediction apparatus 500. Still further, the processor 510 may be configured to communicate with the storage medium 530, and execute a series of instruction operations in the storage medium 530 on the user drunk driving behavior analysis and prediction apparatus 500.
The user drunk-driving behavior analysis and prediction device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input-output interfaces 560, and/or one or more operating systems 531, such as Windows server, Mac OS X, Unix, Linux, FreeBSD, and the like. It will be understood by those skilled in the art that the configuration of the user drunk driving behavior analysis prediction apparatus shown in fig. 5 does not constitute a limitation of the user drunk driving behavior analysis prediction apparatus, and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components.
The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium, and which may also be a volatile computer-readable storage medium, having stored therein instructions, which, when run on a computer, cause the computer to perform the steps of the method for analyzing and predicting drunk driving behavior of a user.
The invention also provides a user drunk driving behavior analysis and prediction device, which comprises a memory and a processor, wherein the memory stores instructions, and the instructions are executed by the processor, so that the processor executes the steps of the user drunk driving behavior analysis and prediction method in each embodiment.
Further, the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the blockchain node, 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.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A method for analyzing and predicting drunk driving behaviors of a user is characterized by comprising the following steps:
receiving a drunk driving behavior analysis request of a user, and acquiring user detail data according to the drunk driving behavior analysis request of the user, wherein the user detail data comprises case work order information, position-based service track data of the user and action hotspot track data of the user;
performing data cleaning on the position-based service track data of the user and the action hotspot track data of the user according to a preset data cleaning rule to obtain cleaned position-based service track data and cleaned action hotspot track data;
performing data analysis and statistical processing on the case work order information, the cleaned location-based service track data and the cleaned action hotspot track data to obtain a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors;
carrying out drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result;
and generating user drunk driving evaluation information according to the drunk driving prediction result, and pushing the user drunk driving evaluation information to a target terminal so that the target terminal processes corresponding vehicle compensation business according to the user drunk driving evaluation information.
2. The method for analyzing and predicting drunk driving behaviors of users according to claim 1, wherein the step of performing data cleaning on the position-based service trajectory data of the users and the action hotspot trajectory data of the users according to a preset data cleaning rule to obtain cleaned position-based service trajectory data and cleaned action hotspot trajectory data comprises the steps of:
performing data analysis on a preset data cleaning rule according to a preset data format to obtain an analyzed data cleaning rule, wherein the analyzed data cleaning rule comprises a track data type, a plurality of track fields and a data cleaning conversion expression corresponding to each track field;
and respectively carrying out data deduplication, abnormal data deletion, null value data cleaning and data format standardization processing on the position-based service track data of the user and the action hotspot track data of the user in sequence based on the track data type, the plurality of track fields and the data cleaning conversion expression corresponding to each track field to obtain cleaned position-based service track data and cleaned action hotspot track data.
3. The method of claim 1, wherein the analyzing and statistically processing the case work order information, the cleaned location-based service trajectory data, and the cleaned action hotspot trajectory data to obtain a risk feature factor dataset, wherein the risk feature factor dataset is used to indicate a plurality of feature factors related to drunk driving behavior, and comprises:
performing data analysis on the case work order information to obtain the case occurrence time and the case occurrence place;
screening the peripheral environment information of the case location according to the case occurrence location and a preset location range, and combining the case occurrence time and the peripheral environment information of the case location into case characteristic factors;
respectively reading a plurality of position service interest point data and a plurality of action hotspot interest point data which are associated with a target place from the cleaned position-based service track data and the cleaned action hotspot track data according to a preset feature extraction strategy, wherein the target place comprises an entertainment place and a dining place;
respectively carrying out clustering analysis on the plurality of position service interest point data and the plurality of action hotspot interest point data according to preset dimensions to obtain position service characteristic factors and action hotspot characteristic factors;
combining the case characteristic factors, the location service characteristic factors and the action hotspot characteristic factors into a risk characteristic factor data set, wherein the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors.
4. The method for analyzing and predicting drunk driving behaviors of users according to claim 1, wherein the step of analyzing and predicting drunk driving on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain drunk driving prediction results comprises the following steps:
inputting the risk characteristic factor data set into a preset drunk driving prediction model, and performing characteristic coding and vectorization processing on the risk characteristic factor data set through a characteristic extraction layer in the drunk driving prediction model to obtain a plurality of embedded vector characteristics, wherein the embedded vector characteristics comprise a position service characteristic vector set and a motion hotspot characteristic vector set;
performing cross feature combination on the plurality of embedded vector features through a factor decomposition machine layer in the drunk driving prediction model to obtain a first output result;
performing characteristic full-connection processing on the plurality of embedded vector characteristics through a deep neural network layer in the drunk driving prediction model to obtain a second output result;
and carrying out drunk driving analysis and prediction on the first output result and the second output result through an activation function layer in the drunk driving prediction model to obtain a drunk driving prediction result, and mapping and storing the drunk driving prediction result and the user drunk driving behavior analysis request into a preset memory data table.
5. The method for analyzing and predicting drunk driving behaviors of a user according to claim 1, wherein the step of generating drunk driving evaluation information of the user according to the drunk driving prediction result and pushing the drunk driving evaluation information of the user to a target terminal so that the target terminal processes corresponding vehicle compensation services according to the drunk driving evaluation information of the user comprises the steps of:
matching a preset information classification table according to the drunk driving prediction result to obtain an information configuration template, and generating user drunk driving evaluation information based on the information configuration template and the case work order information;
the drunk driving evaluation information of the user is pushed to a target terminal through a preset message queue, so that the target terminal processes corresponding vehicle compensation services according to the drunk driving evaluation information of the user, and the vehicle compensation services comprise vehicle damage compensation services and personal injury compensation services.
6. The method of any one of claims 1 to 5, wherein before the receiving the drunk driving behavior analysis request of the user and obtaining user detail data according to the drunk driving behavior analysis request of the user, the user detail data including case work order information, location-based service trajectory data of the user and action hotspot trajectory data of the user, the method further comprises:
extracting position-based service behavior track sample data and action hotspot behavior track sample data from a preset drunk driving refusal claim case and a preset non-drunk driving refusal claim case;
converting the position service-based behavior track sample data and the action hotspot behavior track sample data into position service-based sample characteristic factors and action hotspot sample characteristic factors according to the moment dimension and the space dimension;
and training an initial deep neural network model according to the position service sample characteristic factors and the action hotspot sample characteristic factors to obtain the preset drunk driving prediction model.
7. The method for analyzing and predicting the drunk driving behavior of the user according to any one of claims 1 to 5, wherein after the drunk driving assessment information of the user is generated according to the drunk driving prediction result and is pushed to a target terminal so that the target terminal processes a corresponding vehicle compensation service according to the drunk driving assessment information of the user, the method for analyzing and predicting the drunk driving behavior of the user further comprises the following steps:
constructing a user behavior portrait according to the user detail data, the drunk driving prediction result, the user drunk driving evaluation information and the vehicle compensation service;
and storing the user behavior portrait in a preset graph database, and sending the user behavior portrait to a target terminal.
8. The utility model provides a user's drunk driving behavior analysis prediction unit which characterized in that, user's drunk driving behavior analysis prediction unit includes:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for receiving a drunk driving behavior analysis request of a user and acquiring user detail data according to the drunk driving behavior analysis request of the user, and the user detail data comprises case work order information, position-based service track data of the user and action hotspot track data of the user;
the cleaning module is used for carrying out data cleaning on the position-based service track data of the user and the action hotspot track data of the user according to a preset data cleaning rule to obtain cleaned position-based service track data and cleaned action hotspot track data;
the analysis module is used for carrying out data analysis and statistical processing on the case work order information, the cleaned location-based service track data and the cleaned action hotspot track data to obtain a risk characteristic factor data set, and the risk characteristic factor data set is used for indicating a plurality of characteristic factors related to drunk driving behaviors;
the prediction module is used for carrying out drunk driving analysis and prediction on the risk characteristic factor data set based on a preset drunk driving prediction model to obtain a drunk driving prediction result;
the generation module is used for generating user drunk driving evaluation information according to the drunk driving prediction result and pushing the user drunk driving evaluation information to a target terminal so that the target terminal processes corresponding vehicle compensation business according to the user drunk driving evaluation information.
9. A user drunk driving behavior analysis and prediction device is characterized by comprising: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invokes the instructions in the memory to cause the user drunk driving behavior analysis prediction apparatus to perform the user drunk driving behavior analysis prediction method of any one of claims 1-7.
10. A computer-readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement the method for analyzing and predicting drunk driving behavior of a user according to any one of claims 1-7.
CN202110474075.3A 2021-04-29 2021-04-29 Method, device and equipment for analyzing and predicting drunk driving behaviors of user and storage medium Pending CN113221984A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110474075.3A CN113221984A (en) 2021-04-29 2021-04-29 Method, device and equipment for analyzing and predicting drunk driving behaviors of user and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110474075.3A CN113221984A (en) 2021-04-29 2021-04-29 Method, device and equipment for analyzing and predicting drunk driving behaviors of user and storage medium

Publications (1)

Publication Number Publication Date
CN113221984A true CN113221984A (en) 2021-08-06

Family

ID=77090191

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110474075.3A Pending CN113221984A (en) 2021-04-29 2021-04-29 Method, device and equipment for analyzing and predicting drunk driving behaviors of user and storage medium

Country Status (1)

Country Link
CN (1) CN113221984A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115100861A (en) * 2022-06-22 2022-09-23 公安部交通管理科学研究所 Drunk driving vehicle identification method

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014207558A2 (en) * 2013-06-27 2014-12-31 Scope Technologies Holdings Limited Onboard vehicle accident detection and damage estimation system and method of use
CN110069988A (en) * 2019-01-31 2019-07-30 中国平安财产保险股份有限公司 AI based on multidimensional data drives risk analysis method, server and storage medium
WO2020020088A1 (en) * 2018-07-23 2020-01-30 第四范式（北京）技术有限公司 Neural network model training method and system, and prediction method and system
WO2020035996A1 (en) * 2018-08-17 2020-02-20 ソニー株式会社 Information processing device, information processing system, information processing method, and program
CN111178452A (en) * 2020-01-02 2020-05-19 平安科技（深圳）有限公司 Driving risk identification method, electronic device and readable storage medium
CN111414874A (en) * 2020-03-26 2020-07-14 中国平安财产保险股份有限公司 Driving risk prediction method, device and equipment based on human face and storage medium
CN112035644A (en) * 2020-09-01 2020-12-04 北京四维图新科技股份有限公司 Vehicle insurance scheme adjusting method and device, electronic equipment and storage medium
CN112348403A (en) * 2020-11-26 2021-02-09 德联易控科技(北京)有限公司 Wind control model construction method and device and electronic equipment

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014207558A2 (en) * 2013-06-27 2014-12-31 Scope Technologies Holdings Limited Onboard vehicle accident detection and damage estimation system and method of use
WO2020020088A1 (en) * 2018-07-23 2020-01-30 第四范式（北京）技术有限公司 Neural network model training method and system, and prediction method and system
WO2020035996A1 (en) * 2018-08-17 2020-02-20 ソニー株式会社 Information processing device, information processing system, information processing method, and program
CN110069988A (en) * 2019-01-31 2019-07-30 中国平安财产保险股份有限公司 AI based on multidimensional data drives risk analysis method, server and storage medium
CN111178452A (en) * 2020-01-02 2020-05-19 平安科技（深圳）有限公司 Driving risk identification method, electronic device and readable storage medium
CN111414874A (en) * 2020-03-26 2020-07-14 中国平安财产保险股份有限公司 Driving risk prediction method, device and equipment based on human face and storage medium
CN112035644A (en) * 2020-09-01 2020-12-04 北京四维图新科技股份有限公司 Vehicle insurance scheme adjusting method and device, electronic equipment and storage medium
CN112348403A (en) * 2020-11-26 2021-02-09 德联易控科技(北京)有限公司 Wind control model construction method and device and electronic equipment

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115100861A (en) * 2022-06-22 2022-09-23 公安部交通管理科学研究所 Drunk driving vehicle identification method

Similar Documents

Publication Publication Date Title
CN109284372B (en) User operation behavior analysis method, electronic device and computer readable storage medium
Tayal et al. Efficiency analysis for stochastic dynamic facility layout problem using meta‐heuristic, data envelopment analysis and machine learning
CN113177660B (en) Driving intention prediction and processing method, device, equipment and storage medium
CN115688760B (en) Intelligent diagnosis guiding method, device, equipment and storage medium
CN110647995A (en) Rule training method, device, equipment and storage medium
CN107220902A (en) The cascade scale forecast method of online community network
Wang et al. Time-variant graph classification
CN115098650A (en) Comment information analysis method based on historical data model and related device
CN114138968A (en) Network hotspot mining method, device, equipment and storage medium
CN115170027A (en) Data analysis method, device, equipment and storage medium
CN113221984A (en) Method, device and equipment for analyzing and predicting drunk driving behaviors of user and storage medium
Zubi et al. Using data mining techniques to analyze crime patterns in the libyan national crime data
CN113487241A (en) Method, device, equipment and storage medium for classifying enterprise environment-friendly credit grades
CN116756225B (en) Situation data information processing method based on computer network security
CN111104422B (en) Training method, device, equipment and storage medium of data recommendation model
CN107644042B (en) Software program click rate pre-estimation sorting method and server
CN109597702B (en) Root cause analysis method, device, equipment and storage medium for message bus abnormity
CN115357656A (en) Information processing method and device based on big data and storage medium
CN113254672B (en) Method, system, equipment and readable storage medium for identifying abnormal account
CN113821418B (en) Fault root cause analysis method and device, storage medium and electronic equipment
US11822578B2 (en) Matching machine generated data entries to pattern clusters
Merschmann et al. A lazy data mining approach for protein classification
CN113627997A (en) Data processing method and device, electronic equipment and storage medium
CN114138709A (en) Librarian room integrated management realized based on cloud archive integrated platform
CN110717521A (en) Intelligent service implementation method and device and computer readable storage medium

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