12 results about "Blacklisting" patented technology

The application relates to a user portrait-based account security monitoring method and device and electronic equipment, the method comprising: acquiring user behavior data; determining a clustering gang according to the user behavior data and a preset clustering model; determining a clustering gang score according to a preset clustering gang scoring rule and the clustering gang; determining a risk score according to the user behavior data and a real-time prediction model; acquiring user login data, a preset blacklist and a third-party risk control score; and determining whether to intercept according to the clustering gang score, the risk score, the third-party risk control score, the user login data, the preset blacklist and a preset risk control rule. The application has the effect of improving the real-time performance of an account security monitoring system.

A neural network apparatus for generating a new chemical structure may receive a structure input of a chemical structure; generate, based on the structure input, a negative attention vector that indicates a respective probability of presence of each of a plurality of blacklists in the structure input; generate a structure expression by encoding the structure input; generate a final reverse blacklist vector that does not include the plurality of blacklists, based on the negative attention vector and the structure expression; and generate the new chemical structure by decoding the final reverse blacklist vector.

The application provides a black list confusion method and device based on a secret sharing network and a judgment system. The method comprises the following steps: a confusion node receives secret shares sent by N bank nodes; wherein, the black list data of each bank node is regarded as a secret, the secret is split into N secret shares, and the N secret shares are sent to N confusion nodes respectively, N is a natural number greater than or equal to 3; the confusion node sums and encrypts the received secret shares to obtain confusion data; and the confusion node sends the confusion data to a judgment node; wherein, the confusion data is used by the judgment node to judge whether the secret is a black list. Based on the secret sharing network, the secret splitting algorithm and the secret confusion, the application can prevent brute force cracking, so that the black list customer data of the shared banks can be shared under the premise of ensuring data security and confidentiality, and the interference of the random value of the local differential privacy algorithm is avoided, and the judgment result is more accurate.

The invention discloses a blacklist detection method and device based on face verification and a medium, and relates to the technical field of computer vision and intelligent monitoring. The method comprises the following steps: acquiring video stream data from a plurality of image acquisition devices in a multi-thread parallel manner to obtain a video frame set containing task identification and image data, and performing face detection on each frame of image in the video frame set to position a face area in each frame of image in the video frame set; extracting a feature vector of the face region to obtain an original face feature vector, and performing normalization processing on the original feature vector to obtain a standard face feature vector; and based on a preset task number and the current timestamp, constructing a retrieval condition, and based on the retrieval condition, performing similarity retrieval on the standard face feature vector in a preset vector database to obtain a blacklist result. Through multi-thread parallel acquisition, face detection, feature extraction and vector retrieval, accurate recognition of key targets is realized.

This invention discloses a cross-language code vulnerability remediation method based on dual constraints of knowledge graph and large model, comprising: receiving multi-language heterogeneous vulnerability code samples, performing feature standardization processing, and extracting the language context and core vulnerability intent tags of the code; inputting the vulnerability intent tags into a pre-built cross-language security knowledge graph, performing cross-level pathfinding matching based on risk patterns, and dynamically extracting the necessary security interfaces for vulnerability remediation and prohibited dangerous functions; using a context-aware mechanism, fusing and concatenating the original vulnerability code with whitelist rules to construct prompt words as soft constraint intent guidance at the input end of the large model; simultaneously, mapping blacklist rules to a set of underlying lexical indexes, introducing a log-probability interceptor in the decoding and inference stage of the large model, forcibly modifying the generation log-probability of blacklisted lexical elements to negative infinity, and implementing hard constraint interception at the underlying level; finally, combining the soft guidance at the input end and the hard interception at the decoding end, bidirectionally controlling the autoregressive generation process of the pre-trained language model, thereby eliminating model illusion and outputting the optimal security remediation patch that conforms to the target language specification.

The invention relates to the technical field of sales data statistics, in particular to an intelligent conversion and statistical method and system for business sales data, and the method comprises the steps: collecting original sales data in real time, carrying out the data preprocessing of the original sales data, generating processed sales data, and carrying out the statistical processing of the processed sales data. Associating the processed sales data to a corresponding salesman entity; based on the processed sales data, dynamically generating a sales combat containing key performance indicator visualization; according to a preset dynamic evaluation model, in combination with the associated processed sales data, performing comprehensive performance scoring and ranking on the salesman entities, and determining a red list and a black business list based on a ranking result; and comparing and analyzing the behavior data characteristics of the employee entities in the red list and the black list, and generating different sales strategy suggestions. According to the application, closed-loop management from data acquisition to strategy generation is realized, and the automation and intelligence level and decision-making efficiency of sales management are improved.

This invention discloses an optimized access control system for all types of electronic products based on a dual-layer independent authentication element, belonging to the field of device security authentication technology. The system adopts a dual-layer independent architecture: a hardware-fixed first authentication element and a user-created second authentication element. The first authentication element is the factory-fixed baseline information of the device, while the second authentication element is generated after the user's first verification of a custom sub-element. The system generates an authorization credential uniquely bound to the current device and application, which cannot be reused across devices or applications. For consecutive authentication failures, the system implements tiered delay, temporary blacklisting, and time-limited blacklisting strategies, only rejecting authentication requests without modifying user data. Users can choose whether to clear historical authorization credentials when modifying custom sub-elements. Only those with valid first authentication element verification can access the second authentication data. The system supports authorization record management and automatic cleanup upon uninstallation. This invention features a simple architecture, high security, effective prevention of brute-force attacks and theft, strong versatility, and applicability to all types of electronic products.

The application discloses a WeChat for enterprise customer loss risk early warning method and system, comprising the following steps: S1, acquiring conversation messages, customer behavior events and staff follow-up states in real time through a WeChat for enterprise interface; S2, cleaning and normalizing the collected data, calculating the feature scores of five dimensions of active degree decline rate, response delay growth, customer loss signal, follow-up quality score and risk history accumulation; S3, weighting and summing up the feature scores of each dimension according to preset weights to obtain a comprehensive risk score, and determining a risk level according to the preset interval in which the comprehensive risk score is located; and S4, executing corresponding early warning information pushing and customer management actions according to the risk level. The application belongs to the technical field of conversation management systems, and the structure can realize real-time and accurate capture of customer active deletion and blacklisting behaviors, effectively reducing the false negative rate and the false positive rate.

The invention relates to a customer financial risk grading method and device, equipment, a medium and a product. The method comprises the steps of performing rule scoring on target customer data based on a preset rule engine and at least one risk index to obtain a first risk score value for evaluating dominant risks; performing model scoring on the feature vector of the target customer data based on a trained machine learning model to obtain a second risk score value for evaluating the hidden risk; the machine learning model is obtained by training historical customer data as a sample, and the historical customer data comprises judicial check and control data, blacklist data, suspicious list data and untrusted licensed pedestrian data; performing weighted fusion on the first risk score value and the second risk score value to obtain a final risk score; and mapping the final risk score into a corresponding customer risk level. According to the method, the defects that a pure rule system is high in subjectivity and poor in generalization ability are effectively overcome, and an objective and comprehensive financial risk grading scheme with prediction ability is provided.

The application discloses a black list mining method and device and a computer readable storage medium. The method comprises the following steps: obtaining customer information of a plurality of customers, constructing at least one knowledge graph according to the customer information, and the plurality of customers comprising black list customers and non-black list customers; extracting customer features of each customer according to the knowledge graph, wherein the customer features comprise graph embedding features, graph structure features and customer risk features; training a single classification black list identification model based on the customer features of the plurality of customers; using the trained single classification black list identification model to predict whether a to-be-tested customer belongs to a black list, and updating a historical black list according to a prediction result. The application combines knowledge graphs, feature engineering, machine learning, artificial intelligence and other technical methods to identify hidden black list customers, automatically expand the black list, obtain a new black list, help enterprises to manage business risks and improve risk management capabilities, and reduce manpower.

The invention relates to the technical field of Internet, in particular to a marketing bidding resource adaptive allocation method and system, and the method comprises the steps: receiving a bidding request, and carrying out the analysis of the bidding request, and obtaining an audience identifier, an audience network identifier and an access address identifier; calling a historical bidding log in a preset observation period, and aggregating to obtain a request scale, an occurrence time period, the number of times of short-time window requests and the diversity of access addresses; generating an anomaly judgment feature set which comprises a daily request proportion, occurrence duration, short-time burst and a diversity ratio; carrying out statistical estimation on the features to obtain a threshold set; comparing each identification feature with a threshold value, writing the identification features meeting abnormal conditions into a blacklist, marking other identification features as conventional, and removing the identification features according to an unappearing time limit; if a blacklist is hit in the bidding pre-processing stage, the request is judged to be a non-intention request and marketing bidding resource allocation is reduced or canceled, otherwise, the request is judged to be an intention request; and outputting a judgment and distribution result, and writing back the log to update the blacklist. According to the invention, the problem of unreasonable resource consumption and distribution caused by mixing of non-intentional bidding requests in the marketing bidding process can be solved.