59results about How to "Improve audit accuracy" patented technology

The invention discloses a text content auditing method based on a sensitive word. The text content auditing method comprises the following steps of: receiving a text to be audited, carrying out parsing and word segmentation on the text to be audited, and obtaining all keywords in the text to be audited; according to all keywords, inquiring a preset sensitive word database, and obtaining the sensitive word in the text to be audited, wherein the sensitive word database comprises sensitive words and the synonyms or the homoionyms of the sensitive words; obtaining the co-occurrence keyword of the sensitive word in a preset text length, calculating the violation weight of the sensitive word and the co-occurrence keyword of the sensitive word, and judging whether the violation weight is greater than a preset violation threshold value or not; and if the violation weight is greater than the preset violation threshold value, proving that the text to be audited is a violation text, and otherwise, proving that the text to be audited is a normal text. By use of the text content auditing method, a misjudgment probability is effectively lowered, auditing accuracy is improved, and the text content auditing method has quick reaction capacity for anagrams and net neologisms.

An embodiment of the present invention provides an audio and video content analyzing method and apparatus. The method includes the following steps: a step of acquiring an audio and video file to be processed, a step of acquiring attribute information of an associated user related to the audio and video file to be processed, and a step of performing content analysis on the audio and video file to be processed according to attribute information of the associated user to output category information. Via the audio and video content analyzing method and apparatus disclosed in the technical solutionprovided in the embodiment of the present invention, content review is performed on the processed audio and video file to be processed based on the audio and video file to be processed and the attribute information of the associated user related to the audio and video file to be processed, the attribute information of the associated user includes a large quantity of data related to video contentthat can be utilized and mined; auxiliary classification of videos is performed based on the data, reviewing accuracy can be effectively improved, and advancement of video reviewing automation can befacilitated.

The present application discloses a method and a system for reviewing chat content based on sensitive words. The method for reviewing chat content comprises: dividing a sensitive word vocabulary intoat least two sensitive word sets including jargon sets; matching each of the sensitive word sets with a text library to generate a text set corresponding to each of the sensitive word sets; taking texts in the text sets as training materials, and performing machine learning processing on the training materials to generate a classification review model corresponding to each of the sensitive word sets; and sending to-be-viewed texts into one or more classification review models for review, and obtaining review results. According to the technical scheme of the present application, problems that,by simply comparing keywords, the existing text review system can easily lead to misjudgment and cannot respond to new words and jargon in a timely and effective manner.

The application relates to a vehicle violation video processing method and device, computer equipment and a storage medium. The method comprises steps: acquiring license plate information of a targetvehicle and multiple frames of video images in a vehicle violation video; detecting each frame of video image by a target detection model to obtain position information of each vehicle in each frame of video image; determining the driving direction of the target vehicle according to the license plate information of the target vehicle and the position information of each vehicle in each frame of video image; performing scene segmentation on each frame of video image by a segmentation model to obtain segmentation results corresponding to all frames of video images, carrying out fusion on the segmentation results to determine final scene information, and determining a guide line type of a lane where the target vehicle is located according to the final scene information; determining whether the guide line type matches the driving direction of the target vehicle or not; and if so, determining that the target vehicle does not break traffic regulations. Therefore, whether the target vehicle breaks traffic regulations can be determined by using multiple frames of video images, so that the examination accuracy is improved.

The invention provides a virtual domain-based HTTP reduction display method, which comprises that: an auditing system establishes a virtual domain server for reduction on an HTTP server in advance, and when a user opens a file to be reduced through a browser, the auditing system redirects a request to a virtual domain; the virtual domain server searches a reduction file in the system according to the preset unique identifier contained in the request, and returns the reduction file to the browser for display; and when the browser requests to reduce other segments in the content for the virtual domain server again, the virtual domain server judges a reduction file corresponding to the request through the unique identifier, and returns the reduction file corresponding to the request to the browser. The method can solve the problem of complete display of webpages in the process of auditing and reducing the contents of the webpages.

The invention relates to the technical field of database audit and safety management, and aims at providing a method for improving three-layer correlation accuracy. The method comprises the following steps: obtaining information generated during interaction between a Web server and a front-end accessor and information generated during interaction between the Web server and a background database through a proxy server, and sending the information to a database audit system; and receiving, by the database audit system, the data sent by the proxy server, and carrying out one-to-one correlation to determine the correlation of front-end HTTP access and background database access. According to the method, the efficiency of the one-to-one correlation between the background database access and the front-end accessor is greatly improved, so that the audit efficiency of the database audit system is improved; and meanwhile, the correctness of the one-to-one correlation between the background database access and the front-end accessor is greatly improved, so that the audit correctness of the database audit system is improved.

The invention provides a file auditing method, a file auditing device and a file submission control system. The method comprises the steps of receiving a to-be-submitted file transmitted from a version control system; auditing the to-be-submitted file to obtain an audit result; and feeding the obtained audit result back to the version control system. Through the file auditing method and the file auditing device, the problems that adoption and rejection only can be simply executed when submission is carried out through an interception function of the version control system in related technology and the modifying operation on the interception function is inconvenient are solved. The auditing process is transferred from the version control system, the audit result can be refined and the auditing rule can be conveniently modified, so that the effects of improving the auditing efficiency and improving the auditing accuracy are achieved.

The invention provides an information verification method and device and equipment for risk control, and a storage medium. The method comprises: when business data is received, establishing or updating an information knowledge graph according to a preset data association rule and the business data; judging whether risk data conforming to a preset verification rule exist in the service data or notaccording to an association relationship of the service data in the information knowledge graph; and if the risk data exists in the service data, performing risk identification on the risk data, and generating and displaying a risk reminding message according to the risk identification. According to the method, the information knowledge graph is established according to the incidence relation in the service data, the verification rule is preset according to the data characteristics of the risk data, and the risk intelligent prediction is automatically carried out on the service data accordingto the preset verification rule and the information knowledge graph. Therefore, the problem of low efficiency caused by manual audit is avoided, the audit efficiency is improved, and the audit accuracy is improved.

The application relates to the field of communication technology, and particularly to a service object identification method and device, which are used to solve the problem that accuracy of current manners of auditing service object information is poorer. The service object identification method provided by the embodiment of the application includes: acquiring to-be-audited service object information of a service object provider, wherein the to-be-audited service object information includes a service object identifier and a service object display information used for display to a service object requester; extracting historical data, which correspond to the service object identifier, from a historical database according to the service object identifier, wherein the historical database is used for storing basic data of all service objects in a service platform and related data of service processing; and through comparing the service object display information with the historical data extracted from the historical database, determining whether a to-be-audited service object is an actual service object corresponding to the service object identifier.

The invention relates to the data processing technology field and particularly relates to an insurance business vehicle information audit method and device, a storage medium and electronic equipment.The method comprises steps that wireless communication connection with a to-be-checked vehicle is established, and wireless communication connection is continued in the check process; a request for acquiring an identification code of the to-be-checked vehicle is sent to to the to-be-checked vehicle through wireless communication connection; an identification code of the to-be-checked vehicle returned for responding to the request is received through wireless communication connection; the identification code is sent to the server; the vehicle information which is returned by the server and corresponds to the identification code is received and is for check of the to-be-checked vehicle. The method is advantaged in that uniqueness and validity of the to-be-checked vehicle in the whole check process are guaranteed, the mode is simple and effective, the check step is simple, the method is not limited by location, light and environment factors, user experience is relatively good, moreover, compared with the prior art, check efficiency is higher, and check accuracy is higher.

The invention provides an automobile defect data automatic input method, an automobile defect data automatic input system, an automobile defect data automatic input device and a medium. The automobiledefect data automatic input method comprises the steps of: receiving audio data of a user; performing voice recognition on the audio data, and determining defect information corresponding to the audio data; acquiring a defect region image of a defect position of an automobile; identifying the defect region image, and determining identification information of a defect region; and storing the defect information and the identification information of the defect region. According to the automobile auditing auxiliary scheme provided by the invention, an intelligent auxiliary function is provided for an auditing process by combining voice recognition, image recognition and other technologies, defect information and identification information of the defect region are automatically input, and theauditing efficiency and precision are improved.

The invention discloses a credit customer qualification classification method based on WOE conversion through machine learning. A system comprises a data preparation and preprocessing module, a modeltraining and evaluating module, a model deployment module, an inlet data processing module and a client qualification division module. The data preparation and preprocessing module is used for calculating original data I from the application data, the credit investigation data and the call record, calculating original data II through the customer category and the repayment data, and carrying out data preprocessing on the original data I and the original data II. The invention relates to the technical field of qualification classification. The credit customer qualification classification methodbased on WOE conversion through machine learning provides the system for realizing customer classification with different qualifications based on the machine learning method, the workload of manual auditing can be reduced, the approval efficiency is improved, learning is performed in time according to newly added customer information, self-adaption to customer qualification change is realized, the manual auditing efficiency can be improved to a greater extent, and the labor cost is reduced.

The embodiment of the invention discloses a machine examination method and related products, which are applied to a server, and the method comprises the following steps: obtaining a code input by an examinee and to-be-verified data, and obtaining a plurality of preset identity codes from a preset database; determining a target identity code corresponding to the code based on the plurality of identity codes, and obtaining target verification data corresponding to the target identity code; and executing a matching operation on the to-be-verified data and the target verification data, and when the to-be-verified data and the target verification data are successfully matched, pushing a preset test question to a terminal corresponding to the to-be-verified data. Therefore, by implementing the embodiment of the invention, the efficiency of identity auditing before examination can be improved, the auditing accuracy is improved, and the user experience is improved.

The invention discloses a loan qualification auditing method based on artificial intelligence and related equipment, and relates to the field of artificial intelligence, wherein the method comprises the steps: building a parameter information table; dialoguing with a user by using a dialogue model, and carrying out intention recognition and parameter extraction on the user utterance by using a deep learning model in the dialogue process; if the identified intention is an informing intention, filling extracted parameters into the parameter information table; standardizing the parameters in theparameter information table; and evaluating the loan qualification of the user according to the parameter information table. According to the method, dialogue with the user is carried out through thedialogue model, the labor cost is reduced, the auditing efficiency is improved, intention recognition and parameter extraction are automatically carried out on the user utterance on the basis, the loan qualification is automatically evaluated according to the parameter information table of the user, the problems of misremembering, missing remembering and the like of the user information are avoided, and the auditing accuracy is improved.

The invention provides a video content auditing method and device, a storage medium and a computer device. The method comprises the steps that: a violation score of each frame of picture of a to-be-detected video is obtained; the first N frames of pictures with the highest violation score are determined, wherein N is greater than 1; the violation scores of the first N frames of pictures are inputted into a trained classifier, and the violation scores of the first N frames of pictures are fitted by the classifier, so that the score of the to-be-detected video can be obtained; and whether the to-be-detected video is illegal or not can be judged according to the score of the to-be-detected video. With the method of the embodiment of the invention, the recall rate of violation videos is improved.

The invention relates to a vehicle illegal video processing method and device, computer equipment and a storage medium. The method comprises the steps of acquiring plate number information of a targetvehicle and a multi-frame video image in vehicle illegal video; and detecting each frame video image by a first depth learning model, a second depth learning model and a third depth learning model toobtain a continuous position of the target vehicle, a continuous position of each pedestrian on pedestrian crossing, a type of each pavement marking and a position of each pavement marking. The detail dynamic process of the target vehicle and the pedestrian is detected by the multi-frame image, so that whether the target vehicle is not illegal or not can be more accurately judged, and the vehicleillegal video approval accuracy is improved.

The invention provides a picture auditing method. The picture auditing method comprises the steps of obtaining a to-be-put picture; obtaining attribute information corresponding to the putting position of the picture; based on the attribute information, extracting a plurality of pixel points to be audited and color values of the plurality of pixel points from the picture; and auditing the picturebased on the color difference value of the color values of the plurality of pixel points and a preset color value. The invention further provides a picture auditing device and system.

The present disclosure provides an information auditing method, device, electronic device, and computer storage medium, and the method is applied to a voice loan system which enters load information in a voice way. The method comprises: in the process of entering the loan information by voice, recording reaction time duration of entering the target loan information, wherein the reaction time duration refers to a time interval from the completion of voice broadcast for promoting to enter the target loan information to the beginning of entering the target loan information; auditing the target loan information based on the reaction time duration; and determining a loan amount based on the entered loan information in the case that the auditing of the target loan information passes. In this way, the auditing accuracy of the loan information can be improved, and the credit risk of a credit platform can be reduced.

The invention relates to the field of artificial intelligence and digital medical treatment, is applied to the field of intelligent medical treatment, and discloses a medical data anomaly detection method, device and equipment and a storage medium. The method comprises the following steps: receiving hospitalization medical data, and inputting the hospitalization medical data into a preset judgment engine for verification; if the check of the judgment engine is not passed, determining that the hospitalization medical data is abnormal data; if verification of the judgment engine is passed, taking the hospitalization medical data passing verification of the judgment engine as to-be-processed data, and performing disease name extraction on the to-be-processed data by utilizing a disease theme extraction model; according to the disease name, obtaining an identification model, inputting the to-be-processed data into the identification model for abnormality judgment, and determining whether the to-be-processed data is abnormal data; and when the to-be-processed data is abnormal data, outputting the corresponding to-be-processed data. The invention further relates to a block chain technology, and hospitalization medical data is stored in blockchain. The abnormal data in the hospitalized medical data can be accurately screened out.

The invention discloses an information auditing method and device, and relates to the technical field of computers. One specific embodiment of the method comprises: obtaining to-be-audited bill detailinformation, filtering target attributes belonging to a set memory bank from a target attribute set of the bill detail information, and performing word segmentation processing on the filtered targetattributes according to a pre-constructed word segmentation word bank to obtain a first word segmentation result; starting a multi-thread task, segmenting the responsibility-free list information by using the multi-thread task, and performing parallel word segmentation processing on reference attributes of the segmented responsibility-free list information according to the word segmentation word bank to obtain a second word segmentation result; and calculating the matching degree between the first word segmentation result and the second word segmentation result, determining the responsibility-free amount of the bill detail information according to the matching degree, and updating the filtered target attribute to a memory bank. According to the method, automatic auditing of the bill detailinformation is realized, the manual auditing pressure is reduced, and the auditing speed, auditing accuracy and user experience are improved.

The invention discloses a method and device for updating a database and a method and device for checking a processed object. The method for updating the database is applied to a first server which comprises an intelligent checking system for checking the processed object, wherein the first server is connected to a second server and / or a third server, the second server comprises a high-volume database, the third server comprises a manual checking system, and the intelligent checking system corresponds to a first database comprising M checking rules. The method for updating the database includes the steps that first characteristic data are acquired from the high-volume database, and / or second characteristic data are acquired from the manual checking system; updating data used for updating the M checking rules in the first database are extracted from the first characteristic data and / or the second characteristic data; the M checking rules are updated based on the updating data, and N checking rules are acquired, wherein at least one checking rule in the N checking rules is different from any checking rule in the M checking rules.

The invention discloses a device for auditing themes to be published in a network community. The device comprises a data base, an interface displaying module, a key word matching module, a theme high-brightness displaying module and a content high-brightness displaying module, wherein the data base provides a plurality of key words; the interface displaying module is used for displaying all themes which are not audited and need to be published in an interface; the key word matching module is used for matching all the themes to be published according to each key word; if terms with the same key words are matched, the theme high-brightness displaying module is used for performing high-brightness displaying on the corresponding themes to be published in the interface for displaying all the themes to be published; when the theme to be published is opened, the content high-brightness displaying module is used for performing high-brightness displaying on the corresponding terms in the content of the themes to be published. According to the device disclosed by the invention, inelegant key words are matched, and high-brightness displaying is performed on the theme to be published including the inelegant key words; besides, when the theme to be published is opened, the high-brightness displaying is performed on the corresponding terms in the content of the theme to be published, so that auditing is convenient, the labor force is alleviated, and the auditing accuracy rate is greatly increased.

The embodiment of the invention provides a data extraction method and device, and the method comprises the steps of obtaining a data extraction condition set corresponding to a preset storage space after obtaining a first type of data extraction instruction, the data extraction condition being a condition for filtering the invalid vehicle passing data; obtaining the to-be-audited data meeting thedata extraction condition from the to-be-audited data stored in the first storage device as the first to-be-audited data; and storing the obtained first to-be-audited data in a preset storage space, so that an auditor extracts the data from the preset storage space. By applying the embodiment of the invention, the invalid vehicle passing data is filtered, the workload of the audit personnel is reduced to a certain extent, and the audit efficiency of the audit personnel is improved.

The invention relates to a revenue fund auditing system and a revenue fund auditing method. The system comprises an acquisition module, an integrating module, a first auditing module, a second auditing module and a correction module, wherein the acquisition module is used for acquiring receivable, actually-received fund and turned-over fund; the integrating module is used for integrating the receivable and the actually-received fund, forming table data and sending the table data to a client; the first auditing module is used for first consistency auditing, sending the turned-over fund to the second auditing module in the case of consistency, and forming and sending the turned-over fund with a difference reason to the second auditing module in the case of inconsistency; the second auditing module is used for second consistency auditing, forming table data and sending the table data to the client in the case of consistency, and acquiring turned-over fund correction data and calling the correction module in the case of inconsistency; and the correction module is used for correcting the turned-over fund audited through the second consistency auditing and sending the turned-over fund after correction to the first auditing module. Manpower and human resources consumed for revenue fund auditing can be reduced, the auditing timeliness is strong, the accuracy is high, and the financial risks are low.

The invention relates to an image processing-based invisible correction follow-up management system, comprising: a shooting end and a processing end, the shooting end is used to shoot oral images for follow-up visits, and the processing end is used to review and process the oral images taken by the shooting end and feedback; the processing end is equipped with an upload module, an analysis module, an audit module, a transformation module, a sharpening module, a follow-up visit module and a control module. Oral image resolution, the review module is used for review, the conversion module is used to obtain the number of oral image contour lines, the sharpening module is used for image sharpening processing, the follow-up module is used to store the approved oral images for doctors to view, and the control module Used to control the oral image review process. Therefore, the verification accuracy rate can be effectively improved, and compared with the prior art, the time for manual verification is also saved.

The invention discloses an intelligent credit granting system and method for technical medium and small enterprises, and relates to the technical field of credit granting systems.The intelligent credit granting system comprises a credit granting system platform, wherein the receiving end of the credit granting system platform is in signal connection with a credit granting system client, the output end of the credit granting system platform is in signal connection with a credit granting system auditing end, and the credit granting system auditing end comprises a manual auditing port and an intelligent system auditing port. Through the cooperation of the manual auditing port and the intelligent system auditing port, auditing channels are increased and the auditing accuracy and efficiency are ensured, and through the cooperation of a decryption distribution module and an intelligent analysis module, credit granting material information is intelligently analyzed, the difficulty and errors during auditing are reduced, and the auditing efficiency is improved. According to the system, the problems that errors are easy to occur and the credit granting auditing efficiency is relatively low because most credit granting systems are relatively dependent on a manual mode for auditing are solved, the utilization efficiency of the credit granting system is improved, and the use cost is reduced.