10 results about "Preference behavior" patented technology

The invention relates to the technical field of artificial intelligence, can be applied to business system platforms of financial science and technology, medical health and the like, and discloses a man-machine interaction-based question answer feedback method, device, equipment and medium, the method comprises the following steps: obtaining a question help request of a target user, obtaining a question frequency, and if the question frequency is less than a preset frequency threshold, sending a question answer to the target user; if yes, intention recognition is conducted on the question help-seeking request to obtain question key information, and if yes, context reloading is conducted on an obtained historical question record according to the question help-seeking request to obtain the question key information, statement splitting is conducted on the question key information to obtain a key task; and searching related information of the key task from the obtained external professional knowledge base, generating a user preference behavior according to the historical question record, generating a question answer of the question help request according to the related information, and sending the question answer to a preset target user side. According to the invention, the efficiency of answering various questions for the user is improved.

The invention provides a skill-driven human preference behavior application method and device, and the method comprises the steps: S10, carrying out the skill learning through unsupervised skill pre-training based on unlabeled intelligent agent environment interaction data, obtaining a skill trajectory, and storing the skill trajectory in a buffer area; s20, selecting a to-be-labeled skill track pair in the buffer area through a query selection mechanism, and performing manual labeling according to human preferences; and S30, training a reinforcement learning model based on the manually labeled skill trajectory pair to obtain a control decision model for controlling the intelligent agent. According to the method, the query distinguishability of the intelligent agent control task is remarkably improved, and the labeling noise is reduced; according to the method, the robustness of intelligent agent control is enhanced, and the dependence on labeled data is reduced; in addition, the method is compatible with various skill discovery methods, and the application scene of the intelligent agent control field is expanded.

The utility model relates to a hole punching experiment box body for testing the social preference behavior of a mouse, and provides the hole punching experiment box body for testing the social preference behavior of the mouse, aiming at the problem that the existing device for testing the social preference behavior of the mouse is inaccurate in judgment, the hole punching experiment box body for testing the social preference behavior of the mouse comprises a box main body with a top opening, the box main body is divided into a movable area and an isolation area by the isolation plate, a trigger pricking hole and an observation opening are formed in the isolation plate, a trigger is arranged at the trigger pricking hole and used for sending a touch signal, a movable plate capable of being driven by a motor is arranged at the observation opening, the motor can be started according to the signal sent by the trigger, and a closed area and an interaction area are arranged on the movable plate. The closed area can cover the observation port, when the motor is started to drive the movable plate to move, the closed area at the observation port is switched into the interaction area, the social motivation of the mouse can be digitized, the detection on the social preference behavior driven by the motivation is more scientific, and the social preference back social motivation driving mechanism can be better reflected.

The application provides a user portrait-based matching strategy generation method and related device, and relates to the field of computers. First, a name list including a plurality of job personnel is acquired, and then job record data of each job personnel in a previous job cycle is acquired; then, a user portrait of each job personnel is determined based on the job record data; and a job object matching strategy is determined for each job personnel, the job object matching strategy corresponding to a job behavior type of the job personnel. Since the user portrait represents a job behavior type matched with the corresponding job personnel, the job behavior type reflects a preferred behavior feature, so that the recommended content using the job object matching strategy can be matched with the preferred behavior feature of the job personnel, and user experience is ensured.

The application provides an artificial intelligence-based content recommendation method and device, an electronic device, and a computer readable storage medium. The method comprises: in response to a touch operation of a content presentation interface, obtaining historical operation behaviors of a user account logged into the content presentation interface; determining an account attribute of the user account according to the historical operation behaviors; obtaining account preference behaviors conforming to the account attribute, and performing preference analysis and processing according to the account preference behaviors to obtain preference information of the user account for multiple contents in a content library; determining a content recommendation strategy according to the account attribute and the preference information, and screening out to-be-recommended contents from the content library according to the content recommendation strategy; and updating the content presentation interface according to the screened out to-be-recommended contents. Through the application, accurate content recommendation can be achieved, and the actual utilization rate of the computing resources consumed by the electronic device when performing content recommendation can be improved.

The invention relates to the technical field of crowd doctor-seeing preference prediction, and discloses a traditional Chinese medicine preference behavior prediction system and method for middle-aged and elderly stroke patients, and the method comprises the steps: collecting initial traditional Chinese medicine treatment behavior data of a source crowd, carrying out the preprocessing, feature extraction and dimension reduction processing of the initial traditional Chinese medicine treatment behavior data, obtaining dimension-reduced traditional Chinese medicine characteristic data; training the deep learning network model and the transfer learning network model by using the dimension-reduced traditional Chinese medicine feature data, performing model integration after training is completed, and generating a traditional Chinese medicine preference prediction model; and inputting clinical data of a patient into the traditional Chinese medicine preference prediction model to obtain a traditional Chinese medicine treatment preference probability. Clinical data of a patient is input into a traditional Chinese medicine preference prediction model, and patient personalized preference weights including traditional Chinese medicine category preference probability and the like are output.

The present application relates to the technical field of enterprise training management and personalized recommendation, in particular, the present application relates to a talent performance dynamic evaluation and training resource intelligent pushing method, the present application collects staff daily task data and active retrieval text and time stamp, adopts the dynamic model of the time decay coefficient of the forgetting curve to calculate the current ability evaluation of each skill, and generates the ability gap label set by comparing with the post benchmark, obtains the intention feature by encoding the semantic of the retrieval text, combines the interactive operation depth to calculate the active inclination intensity index, generates the learning time period preference vector according to the retrieval time period distribution, constitutes the active learning preference behavior fingerprint, matches the candidate resource by the gap label set and determines the content matching degree, weights and fuses the content matching degree, the active inclination intensity index and the time period coincidence degree of the decision time period and the preference vector, obtains the pushing priority evaluation value, reorders the resource after the value, and pushes the resource, realizes the fusion of the ability dynamic evaluation and the learning preference.

The invention relates to the technical field of human resource information management, and discloses an enterprise employee welfare gift bag automatic generation method and system, and the method comprises the steps: collecting the multi-source welfare preference behavior data of an employee, and constructing and dynamically updating an employee welfare preference knowledge graph in combination with the family life demand information of the employee; based on the knowledge graph, factors such as employee occupational features and family structures are integrated to generate employee welfare demand portraits; constructing a welfare strategy decision engine, integrating enterprise budget, supply chain state, market price and employee satisfaction information, and generating a globally optimal welfare configuration scheme; and dynamically adjusting a recommendation strategy based on employee selection behaviors and satisfaction feedback through an employee and commodity interaction model comprising preference prediction and commodity recommendation sub-models at key nodes of the employee birthday and the entry commemorative day, and generating a customized welfare gift bag. According to the invention, the welfare configuration is precise, intelligent and personalized, and the employee satisfaction and the enterprise management efficiency are effectively improved.

The invention discloses a medicine for relieving conditional position preference behaviors of morphine-dependent mice and application of the medicine. The medicine comprises polypeptide and siRNA, and the amino acid sequence of the polypeptide is SCSRRLLAAALLPWCCAAW; the siRNA is a double strand, the sequence of a sense strand of the siRNA is UGUCAAAGUCAACAUGGUGdTdT, and the sequence of an antisense strand of the siRNA is CACCAUGUUGACUGdTdT. According to the application disclosed by the invention, by discovering the specific association between the Ppt1 gene in the anterior clasp cortex and morphine dependence and developing a Ppt1-targeted polypeptide (Ppt1-4-22) and siRNA intervention tool, not only is key theoretical inspiration provided for clinical research on an opioid drug-dependent molecular mechanism, but also an accurate theoretical basis and an effective molecular target can be provided for drug research and development.