42results about How to "Accurate portrait" patented technology

The invention discloses a social media text-based big five personality prediction method. The method comprises the four steps of A, performing preprocessing; B, extracting user features; C, building and training a big five personality prediction model; and D, predicting big five personality of a user. The big five personality preference of the user is mined based on a machine learning method and text information released in social media by the user, so that an accurate user portrait is realized and a technical foundation is laid for personalized search and recommendation.

The invention discloses a warehouse receipt pledge financing management system and method based on a block chain in the supply chain industry. A protocol layer, a data layer, a function layer and a service layer which are fully covered by a warehouse receipt pledge financing service system based on a block chain are designed. A practical Byzantine fault-tolerant node consensus mechanism, a data distributed evidence storage feature, an information transparency feature, a distributed operation and maintenance feature and the like adopted by a protocol layer based on a block chain are more convenient for big data structured analysis and processing, relatively accurate enterprise portraits are realized, and real-time data synchronization and data security are realized. Therefore, unified business decision optimization arrangement and risk control are facilitated, and convenience is provided for penetration type supervision. The system and the method can completely cover the links of warehouse receipt pledge article value dynamic evaluation, pledge release, credit extension and examination, financing subject credit risk control, penetration type supervision and the like, and avoid the problems of repeated warehouse receipt pledge, serious false warehouse receipt pledge phenomenon and low credit extension safety evaluation quality of financial institutions.

The invention discloses a marketing method of a big data intelligent marketing system, the main steps of which are as follows: S101: through the analysis of population attribute characteristics, orientate the gender, age, and education labels of the target group; S102: through the analysis of user state characteristics, orientate the target group Marriage, consumption and residence labels; S103: Through terminal device feature analysis, screen the device type, model, and system label of the target group; S104: Through the connection between the offline scene data collection device and the user, determine the user's time, location, and weather and other LBS information; S105: Use the TGI algorithm to analyze the characteristics of user interests and behavior preferences, quantify the degree of user preference, and make three-dimensional user portraits. The present invention quantifies the user's offline behavior by collecting user behavior data and offline hardware device data. The combination of user online data and offline data makes the prediction of user behavior more accurate, and produces marketing content around customer needs.

The invention discloses an account receivable financing management system and method based on a block chain in the supply chain industry. A protocol layer, a data layer, a function layer and a servicelayer which are fully covered by an account receivable financing service system based on a block chain are designed. The protocol layer based on the block chain adopts a Byzantium fault-tolerant nodeconsensus mechanism, the data distributed storage characteristics, the information transparency characteristics, the distributed operation and maintenance characteristics and the like, is more convenient for the big data structured analysis and processing, realizes a relatively accurate enterprise portrait, realizes the real-time data synchronization and data security, and is convenient for the unified business decision optimization arrangement and risk control, at the same time, provides the convenience for the penetration type supervision. By the characteristics of data authenticity storage, tamper resistance, P2P protocol, smart contract, asymmetric encryption and the like of the system disclosed by the invention, the guarantee can be provided for the authenticity of the trade backgrounds of buyers and sellers, the authenticity and realizability of accounts receivable, the rationality and validity of the circulation of the accounts receivable, and the account ages, account periods,even account quality and the like of the accounts receivable.

The invention provides an advertisement putting system based on a block chain, comprising an advertisement release incentive module, an advertisement putting platform module, an advertisement medium module, a user management module and an information verification module. By adopting the block chain technology, the invention can realize the accurate portrait for the user, the data of the block chain is compared by the multi-dimensional difference, the data quality is high, the portrait of the user character is more accurate, the put advertisement contains the content that the user is interestedin, and the advertisement is not prone to make the user feel disgusted. According to the portrait of the user, only the advertisement interested by each user is reserved in the block chain, and the probability of advertisement error is reduced; Data is stored in the network, Only users can access their own complete data, the data is not accessed unauthorizedly, data security is high, advertisingmedia owners disseminate advertising through advertising media, and when clicking and watching advertising through the terminal, the users receive different levels of virtual currency rewards, this way may encourage advertising media owners and users to take the initiative to promote.

the invention discloses a method for user portray In the case of no label in the original training data, a user is classified by using the high heuristic of application name and the feature selectionalgorithm, and then the label information of the training data set is obtained by iterating a classification result repeatedly so that labeled training data set is constructed, and the more accurateuser portrait is finally realized. The invention discloses a user portary device.

The invention discloses a knowledge recommendation method and system based on big data and deep learning. The method comprises the following steps that: according to scene data information, carrying out scene analysis; adopting a deep learning method to process operation and maintenance personnel data information to generate an operation and maintenance personnel portrait; according to a screen analysis result and the operation and maintenance personnel portrait, carrying out knowledge selection to obtain a knowledge recommendation set; and according to the knowledge recommendation set, carrying out knowledge recommendation. The system comprises a scene analysis module, an operation and maintenance personnel portrait generation module, a knowledge selection module and a knowledge recommendation module. By use of the method and the system, the scene analysis result and the operation and maintenance personnel portrait are synthesized to carry out knowledge selection and knowledge recommendation, and knowledge recommendation accuracy is improved from two dimensions including a use scene and a use user; and the deep learning method is adopted to learn data information including the operation behavior of the operation and maintenance personnel and the like, and therefore, the operation and maintenance personnel portrait is more and more accurate. The method and the system can be widely applied to the field of computer application.

The invention is applicable to the technical field of information processing, and provides a social network information management method, terminal equipment and a medium, and the method comprises thesteps: obtaining a contact list of a user according to a received social network information management instruction, the contact list comprising account identifiers of a plurality of contacts; Obtaining a user preference label about a preset social network platform; Respectively collecting social network information associated with each account identifier in the social network platform, and screening out the social network information matched with the user preference tags; And pushing the screened social network information to a user terminal. According to the invention, the social network information conforming to the user preference label is only pushed to the user terminal, so that the user does not need to waste time to check the friend circle information which does not conform to theown needs and does not need to worry about missing and missing the contents which the user needs to pay attention to, thereby reducing the management difficulty of the friend circle information and meeting the personalized needs of the user.

The embodiment of the invention discloses an information recommendation method and device and a server. The method comprises the steps: analyzing retrieval information input by a user, obtaining target reference information matched with the retrieval information, and enabling the similarity between the retrieval information and the target reference information to be greater than a preset similarity threshold; Obtaining historical calling times of the target reference information, wherein the historical calling times are times of matching the target reference information with retrieval information input by at least one user within a first preset time period; Determining recommendation information according to the historical calling times; And sending the recommendation information to the client. By implementing the method, the content which the user may be interested in can be analyzed by combining the historical statistical data, so that more accurate portraying is performed on the user, and the accuracy of the recommendation result is improved.

The invention provides a data processing method and device, computing equipment and a medium, the data processing method is used for analyzing a network attacker, and the method comprises the following steps: obtaining attack alarm information; Obtaining associated information of an attacker corresponding to the attack alarm information based on the attack alarm information; Acquiring feature information of the associated information of the attacker; And obtaining the portrait of the attacker based on the feature information of the associated information of the attacker.

The invention discloses a highway user portrait data acquisition and analysis method and system. A user portrait technology is used, an ETC user portrait construction scheme model is established, an ETC user portrait analysis prototype system is developed, labels such as user basic attributes, user consumption, user travel space-time habits, user value, user activity and user credit are constructed for highway ETC users, and the users are described and depicted from multiple angles. Travel habits of different users on the basis of user grouping are described and further relevance, difference and underlying reasons of passenger car and truck travel are analyzed. From the perspective of time, influences of holidays, days and nights and the like on user travel are analyzed. The system is designed based on the proposed ETC user portrait model, an SQL database and a python language. The invention has the advantages of being efficient in use, low in operation cost, high in result real-time performance and the like, and the actual application requirements can be met.

The invention provides a fingerprint identification method of an industrial control scanner based on honeynet data, comprising the following steps: scanning data captured through a honeypot network system in an industrial control network and an existing industrial control scanner are analyzed to obtain fingerprint information and a multi-classifier model of scanning data based on a CART decision tree is constructed. The multi-classification classifier model can effectively identify the specific scanning tools that initiate the scanning traffic and output the judgment probability of various scanner tags. After that, the output of the multi-classification classifier model will be used as the input data of the clustering algorithm. The clustering algorithm can discover the deeper relationshipbetween different scanning entities and form clustering. At the same time, the clustering algorithm can effectively extract scanning features of different clusters, form new scanner tags, and updatethem to the previous multi-classification decision tree, thereby improving the judgment ability of the invention for the novel scanner data.

The invention discloses a customer relationship management system based on an intelligent analysis technology, which belongs to the technical field of intelligent analysis and comprises a customer information module, a static analysis module, a dynamic analysis module, a matching training module and a dynamic prompt module. The customer information module is used for collecting static information and dynamic information of a customer, wherein the static information comprises the field, the business range, the net profit and the total number of employees of the customer, and the dynamic information comprises contact data, development data and event data of the customer. The static analysis module obtains and updates a static portrait of the customer in real time according to the static information to obtain a first portrait set. The dynamic analysis module obtains and updates a dynamic portrait of the customer in real time according to the dynamic information to obtain a second portrait set. The system is used for solving the technical problem of poor customer relationship management effect in the existing scheme.

The invention provides an internet access behavior recording method and apparatus, and a user terminal. The method comprises the steps of detecting whether a website is a predefined vertical website or not when the website is opened in a browser, wherein the vertical website is a website providing specific field information or related services; when the website opened in the browser is the vertical website, searching for an identification rule matched with the vertical website in identification rules pre-stored in the user terminal; according to the identification rule, identifying and recording data generated by operation behaviors of a user in the vertical website, and according to the data generated by the operation behaviors and user identity information of the user terminal, generating an internet access behavior recording file in a preset format; and sending the internet access behavior recording file to a server for performing user portrait drawing. The data capable of accurately reflecting real interests and hobbies or behavior habits of the user can be collected, so that the user can be subjected to accurate user portrait drawing.

The invention discloses a processing method. The invention discloses a device, equipment and a readable storage medium. The method comprises the steps of obtaining a node association graph and a firstparameter, if the first parameter represents that the node association graph comprises nodes corresponding to different users respectively, if yes, finding a key node from the node association graph;wherein the key node is connected with the first node and the at least one second node, the probability that the key node and the at least one second node belong to different users in the nodes contained in the node association graph is maximum, and cutting off edges between the key node and the at least one second node to obtain at least one sub association graph; According to the method and thedevice provided by the invention, one sub-association graph only comprises the node corresponding to one user, or compared with the prior art, the number of different users corresponding to one sub-association graph is smaller than the number of different users corresponding to the node association graph obtained at present, so that the obtained user portrait is more accurate than that in the prior art.

The invention provides a data processing method and device, equipment and a machine readable storage medium. The method comprises the following steps: receiving a security alarm log; classifying the security alarm logs associated with the same attack source IP into a group, and calculating attack features associated with the attack source IP associated with the group of security alarm logs by a preset rule according to the security alarm logs of the same group; and carrying out clustering calculation according to the attack features associated with each attack source IP, and obtaining an attacker portrait according to a calculation result. Through the technical scheme disclosed by the invention, the security alarm reported by the security equipment is analyzed according to the attack event of each attack source, the attack behavior characteristics of the attack source on a single target are extracted, then the attack behaviors of the attack source on a plurality of attack targets are calculated and analyzed, the attack behavior characteristics of the attack source are extracted, and on the basis, the attack sources are classified by utilizing clustering calculation, the attacker portrait is finally obtained, the efficiency is high, and the portrait is accurate.

The invention discloses a big data comprehensive management method and system for community recreational and sports activities and commercial operation. According to the method provided by the invention, fragmented matches and activities are integrated to form mass flow; through a coupon advertisement putting mode, direct economic income is generated, sponsors can be provided for organizers, relevant data needed by community literature activities and commercial operation are collected in the process, the sponsors are issued to the organizers through the platform, the problem that medium and small activities are difficult to obtain sponsors is solved, and the operation cost of relevant industries is reduced.

The invention discloses a user statistics method and system, and the method comprises the steps: receiving a first broadcast message broadcasted by a user node in a block chain, and obtaining user information carried by the first broadcast message; determining registered geographic position information corresponding to the user information; broadcasting a second broadcast message including the registered geographical location information in the block chain; receiving a fourth broadcast message broadcasted by the data processing node, and obtaining a geographic position information consistencyverification result carried in the fourth broadcast message, the geographic position information consistency verification result being determined by the data processing node according to actual geographic position information and registered geographic position information corresponding to the user information; and counting the users according to the geographic position information consistency verification result. According to the method, the phenomenon that false geographic position information is provided when a user registers information and false information is provided for the live broadcast platform node when false comment information, reward information and other behaviors are published can be avoided, so that the live broadcast platform node can count real user data and can performaccurate user portraying.

The embodiment of the invention provides a data processing method and a device. The method comprises the steps of determining one or more users with regular movement tracks in a to-be-classified userset; determining one or more target users and service preferences of the one or more target users from the one or more users according to the service information related to the one or more users; anddetermining recommended content according to the service preferences of the one or more target users. According to the method of the invention, one or more users with regular activity tracks are determined in the to-be-classified user set, one or more target users are determined according to the service information related to the one or more users, and the recommendation content is determined according to the service preferences of the target users, so that more accurate portraying, classification and recommendation are carried out on the users.

The invention provides information identification management equipment of a Wifi network terminal. The information identification management equipment comprises a circuit board which is provided with a wifiCPU, an onboard antenna, multiple network ports, a USB power supply interface and multiple USB redundant jacks. The onboard antenna is connected with the wifiCPU via a FLASH memory. The network ports are connected with the internal memory of the wifiCPU via a network port filter. The USB power supply interface is connected with the internal memory of the wifiCPU through a clock control chip and a single-chip microcomputer in turn. The clock control chip is respectively connected with the network port filter and a battery arranged on the circuit board. The battery is connected with the internal memory of the wifiCPU. Each USB redundant jack is connected with the wifiCPU through a USB control chip and the FLASH memory in turn. The invention provides the information identification management equipment of the Wifi network terminal. Single or multiple customer flow data can be viewed by users in real time via a mobile phone through wifi terminal scanning so that a basis is provided for offline big data, and accurate image portraying can be performed on the identified users through comparison of multiple pieces of information.

The invention discloses a one-stop full-screen precision programmed data management method, which comprises a one-stop platform and a read identification module, a memory, and a data transceiver. Theobject of the invention is to adopt a one-stop platform, identify and classify the various procedural purchase platforms, the TV end software platform, the software platform of the mobile terminal andthe software platform of the PC terminal carry out accurate identification and classification, with smart TV and mobile phones, the operation of a computer PC at different times under the same IP, Through algorithm analysis, data matching, comprehensive, detailed and accurate target user attributes are obtained, and the obtained data are stored in the data memory, so as to make the advertisementplacement get the best exposure effect. After accurate identification, accurate user portrait is obtained by means of television viewing and recording, television navigation operation and so on.

The invention provides a financial data automatic testing and monitoring system based on a data warehouse. The automatic test system comprises a financial data input module, a financial data preprocessing module, a financial data storage bin, a financial data test engine and a financial data separation bin. A financial data quantization coding module included in the financial data storage bin performs quantization coding on the financial data preprocessed by the financial data preprocessing module to obtain a quantization coding vector of the financial data, and stores the quantization codingvector in the financial data storage bin; and based on the financial data similarity and stability test result of the financial data test engine, the financial data is separated and the separated financial data is stored in the financial data separation bin. The automatic monitoring system monitors whether data exists in the financial data separation bin in the financial data automatic test systembased on the data warehouse or not; and if so, generating a customer portrait based on the data.

The invention discloses a user portrait construction method, and the method comprises the steps: obtaining WIFI feature data, and forming a feature database; identifying a first feature and a second feature of the feature database, the first feature including similarity of the feature data, and the second feature including scene information of the feature data; and constructing a user portrait based on the preset training model, the first feature and the second feature. The invention further discloses a construction device, a terminal and a nonvolatile computer readable storage medium. According to the construction method provided by the invention, the user portrait can be accurately constructed by carrying out statistical analysis on the acquired WIFI feature data.

Embodiments of the present invention provide a data processing method and device. The method includes: determining one or more users with regular activity trajectories in a set of users to be classified; One or more target users and service preferences of the one or more target users are determined among the multiple users; the recommended content is determined according to the service preferences of the one or more target users. In this embodiment of the present invention, one or more users with regular activity trajectories are determined in the set of users to be classified, and one or more target users are determined according to business information related to the one or more users. business preferences to determine the recommended content, so as to achieve more accurate profiling, classification and recommendation of users.

The invention belongs to the field of big data processing, and particularly relates to a creation system of an advertisement analysis database. The creation system comprises a historical user query module, an advertisement feature data extraction module, a user feedback data extraction module, a face recognition module, an image recognition module, a voice recognition module, a video action recognition module, a user tag establishment module, a recognition degree evaluation value calculation module and a database creation module. According to the creation system of the advertisement analysis database provided by the invention, the preferences of the users can be analyzed according to the identity characteristics of the users and the feedback of different advertisements, and then the database containing the identity characteristics, preferences and disgust characteristics of the users is established; and the preference features and the disgust features are extracted from the keyword data set of the advertisement, so that the method is very suitable for the user to depict the consumption demand and consumption information of the user, and the problems that the difficulty of user behavior analysis and interest prediction is high, and accurate portraying cannot be performed on the user can be solved.

The invention discloses a handwritten signature verification method and device assisted by personality personalization, and belongs to the field of human-computer interaction. According to the method, for collected user handwritten signature samples, different classifiers are selected for classification by using different features, and due to the fact that information provided by image features, track features and user character features is different, the discrimination precision can be improved by using diversity of different features, and the discrimination efficiency is improved. Meanwhile, the character of the user can be predicted according to the handwritten signature track information of the user, and the signature of the user is personalized. A support vector machine classifier is trained in a supervised manner through data with positive and negative sample labels and character labels, and the verification accuracy can reach 96%.

The invention provides a user feature extraction method and device, electronic equipment and a readable storage medium. The method is applied to the technical field of information processing. The method comprises the following steps: obtaining application name information, and determining the user characteristics based on the application name information, which are carried out on the terminal equipment side. The risk of user information leakage caused by uploading the application name to the server is avoided; moreover, the user feature information is obtained based on weighted aggregation ofat least two application feature vectors, and even if the extracted user features are uploaded to the server, the application information of the user cannot be reversely deduced based on the uploadeduser features, so that the security of the user information is further ensured.

The invention provides information identification management equipment of a Wifi network terminal. The information identification management equipment comprises a circuit board which is provided with a wifiCPU, an onboard antenna, multiple network ports, a USB power supply interface and multiple USB redundant jacks. The onboard antenna is connected with the wifiCPU via a FLASH memory. The network ports are connected with the internal memory of the wifiCPU via a network port filter. The USB power supply interface is connected with the internal memory of the wifiCPU through a clock control chip and a single-chip microcomputer in turn. The clock control chip is respectively connected with the network port filter and a battery arranged on the circuit board. The battery is connected with the internal memory of the wifiCPU. Each USB redundant jack is connected with the wifiCPU through a USB control chip and the FLASH memory in turn. The invention provides the information identification management equipment of the Wifi network terminal. Single or multiple customer flow data can be viewed by users in real time via a mobile phone through wifi terminal scanning so that a basis is provided for offline big data, and accurate image portraying can be performed on the identified users through comparison of multiple pieces of information.