88 results about "Item Collection" patented technology

An online system automatically generates a personalized collection of items around a theme. The online system generates a prompt for input into a language model, the prompt including information about a plurality of items and a text describing the theme around which the collection of items will be built. The online system requests the language model to generate, based on the prompt, a list of products eligible for building the collection of items. The online system accesses a computer model trained to identify a set of items personalized for a user of the online system. The computer model identifies, based on the list of eligible products and information about the user, the set of items for populating the collection of items. The online system causes a device of the user to display a user interface with the collection of items for inclusion into a cart of the user.

The invention discloses an emergency patient intelligent pre-examination triage system, and particularly relates to the field of emergency patient pre-examination triage, comprising: a data processing module performs standardization processing on patient chief complaint text and sign data, and constructs a pre-examination data set; the disease prediction module inputs the pre-examination data set into a multi-label disease prediction model, and outputs a potential disease set and a prediction probability thereof; the probability adjustment module performs confidence correction on the prediction result in combination with the historical medical history of the patient; the graph construction module constructs a potential disease semantic adjacency graph according to the semantic similarity; the detection item determination module performs co-occurrence path compression on the basis of the semantic graph structure, extracts a minimum detection item set and generates an auxiliary detection item list; the triage guidance module schedules and ranks the detection items according to the real-time states of the detection resources, triage detection guidance is generated, and efficient and flexible emergency treatment primary screening planning is achieved.

The invention relates to the technical field of recording and broadcasting teaching, and discloses a recording and broadcasting teaching course quality evaluation method based on cloud computing. The method comprises five steps of data item collection, data preprocessing, evaluation model construction, evaluation strategy optimization and quality optimization. Wherein the evaluation model construction adopts a dynamic heterogeneous improved graph calculation method, and evaluation reference data is obtained through the steps of construction of an evaluation heterogeneous graph, dynamic updating of a graph structure, construction of a three-layer layering weight mechanism and the like; the evaluation strategy optimization adopts a multi-objective guide improved multi-layer strategy optimization method, and optimal strategy reference data is obtained through the steps of improving a state parameter space, modeling a multi-layer strategy network and the like; and finally obtaining a course quality comprehensive optimization reference scheme. According to the method, multi-modal dynamic data can be comprehensively processed, correlation of each factor is deeply mined, multi-objective optimization is balanced, evaluation accuracy and effectiveness are improved, and specific guidance is provided for quality improvement of recording and broadcasting courses.

The invention provides a cross-modal collaborative recommendation method and system based on a large language model. The method comprises the steps of obtaining identity information of a user and a to-be-recommended item set; inputting the identity information and the item set into a preset recommendation model, so that the recommendation model predicts the click rate of each item in the item set according to the identity information, and outputting an item recommendation list according to the click rate of each item; wherein the recommendation model is constructed on the basis of a large language model and is obtained through training by using a training data set, and the training data set is based on the large language model and a collaborative filtering technology. And the text data of the recommendation fields and the user-project interaction data are constructed after prompt sequence conversion, feature mapping and feature fusion, so that the project recommendation accuracy of the large language model is improved.

In various examples, session-based recommender model systems and applications are disclosed. Systems and methods are disclosed that use a cosine similarity loss during the training of a machine learning model to train the model to generate an item recommendation based on predicting a next item from a sequence of prior items selected within a session. A recommendation model is trained based on training data that represent an ordered sequence of user interactions with the set of items. A set of item embeddings is generated for the set of items. The recommendation model is trained to predict a session embedding that represents a user behavior pattern from a sequence of item embeddings. A cosine similarity loss computed from the session embedding and the item embeddings is used to train the recommendation model. The cosine similarity loss may include both positive and negative cosine similarity components.

An online system receives a request to access an interactive geographical map of sources from a client device associated with a user. The system retrieves data describing a geographical location associated with the user. The system identifies one or more sources within a threshold distance of the location and retrieves data including information describing items available at each source. For each source, the system accesses and applies a machine-learning model to predict a user engagement score indicating a likelihood of one or more interactions by the user with a set of items available at the source if the source is included in the map. Based on the score for each source, the system selects a set of sources and generates the map, in which the map indicates the geographical location of each selected source. The map is then sent to the client device, causing the device to display the map.

The present invention relates to the technical field of intelligent logistics management and control. Disclosed are a human-machine collaborative management and control method, system and device involving logistics storage and transportation. The method comprises: on the basis of a logistics storage and transportation task, determining a target operation process and a corresponding target operation item queue; executing said task on the basis of the queue; determining whether the queue belongs to a manual operation item set; if the queue belongs to the manual operation item set, generating and issuing a manual operation notification instruction; determining whether a confirmation signal fed back on the basis of the manual operation notification instruction is received; if yes, updating the state of operation items to 'in progress'; determining whether the queue belongs to a feedback operation item set; if the queue belongs to the feedback operation item set, determining whether a manual feedback completion signal is received; if yes, updating the state of the operation items to 'completed'; and when the label of one of the operation items is the maximum label in the queue, confirming that said task is completed. According to the present invention, an instruction is sent to an operator and a manual feedback signal is received in time, so as to collaboratively manage and control the logistics storage and transportation operation process requiring manual participation.

An online concierge system receives an order including one or more items from a client device associated with a user and retrieves user data for the user. The system accesses and applies a machine-learning model to predict a measure of preference of the user associated with each item category associated with the items based on the user data and identifies an item associated with at least a threshold predicted measure of preference. The system retrieves picker data for a picker assigned to collect the items and predicts a level of expertise of the picker associated with collecting the identified item based on the picker data. If the predicted level of expertise is less than a threshold, the system identifies an expert picker based on picker data for the expert picker and sends a prompt to a client device associated with the expert picker to assist with collecting the identified item.

The invention belongs to the technical field of data processing. The invention provides a complete-set material management and control method and system and computer equipment, and the method comprises the steps: building a bidirectional correlation index of a suite and subitems, and carrying out the correlation display of the to-be-purchased suite and the subitems according to the bidirectional correlation index when a suite purchasing application is received; according to received purchase order arrival information, generating arrival traceability codes of arrival subitems; according to a received suite receiving application, determining subitems of the suite according to the bidirectional association index, and according to traceability codes of the determined subitems and an approved receiving application form, determining a to-be-delivered suite and updating a suite inventory; and when a sub-item receiving application is received, according to the arrival traceability code of the sub-item and the approved receiving application form, determining the sub-item to be delivered and updating the sub-item inventory at the same time, and according to the bidirectional association index and the updated sub-item inventory, updating the kit inventory, thereby improving the inventory utilization efficiency and the receiving accuracy.

An example method includes receiving a request to identify a set of media items for playback to a user. The method further includes providing information about the request to a diffusion model (DM) component and receiving, from the DM component, a set of vectors corresponding to the information about the request. The method also includes selecting, using a different component, a set of media items based on the set of vectors, and presenting information about the set of media items to the user.

The invention discloses a page interaction method and device, equipment, a medium and a product. The method comprises the following steps: displaying an information board of a live broadcast item on a live broadcast page; wherein when the inventory of the selected supply site of the live broadcast article does not meet the supply condition, information used for prompting switching of the supply site is displayed in a decision interaction area in the information board; in response to a trigger operation in the decision interaction area, displaying an article collection interface in the live broadcast page, and displaying a site switching pop-up window on the upper layer of the article collection interface; wherein an article set corresponding to the selected supply site is displayed in the article set interface, and the site switching pop-up window is used for prompting to switch to a target supply site of which the stock of live broadcast articles meets the supply condition; and in response to the confirmation operation of the station switching pop-up window, switching and displaying the article set of the target supply station in the article set interface, so that flexible matching of supply and demand resources is realized, and balanced distribution of the supply and demand resources is ensured.

Disclosed embodiments provide systems and methods related to collecting return items using an automated kiosk based on a real time risk decision. The automated kiosk captures return item information representing a return item and transmits the return item information and a request for return risk level relating to the return item to a server operable to execute a machine learning model trained on historical information to determine the risk level. The server determines the risk level based on the received return by using the machine learning model and transmits the determined risk level to the kiosk in real-time. Based on the determined risk level and a return amount associated with the return item, the server may also process a refund in real-time.

Systems and methods are described for disambiguating user input based on a physical location of items in a vicinity of a user. The system determines that a query received from a user contains an ambiguity. In response, the system identifies several items in the physical vicinity of the user. Then, the system analyzes the identified plurality of items to determine whether the plurality of items forms a first ensemble of items or a second ensemble of items. If the plurality of items forms a first ensemble of items, the system performs a search using the search query and a first keyword related to the first ensemble of items. If the plurality of items forms a second ensemble of items, the system performs a search using the search query and a second keyword related to the second ensemble of items. The system then outputs results of the performed search.

The embodiment of the invention relates to a to-be-detected item typesetting method and device, electronic equipment and a storage medium, and the method comprises the steps: under the condition that a target set type is an enzyme immunoassay item set, according to preset enzyme immunoassay strategy information and item names corresponding to all to-be-detected items, setting the to-be-detected items according to the preset enzyme immunoassay strategy information; performing layout sorting on each experiment number corresponding to the plurality of items to be detected, and adding a fixed position before the first experiment number and / or after the last experiment number of each item to be detected, so as to generate a first layout file suitable for the microplate reader; under the condition that the target set type is a fixed electrophoresis type item set, according to preset electrophoresis typesetting information, item names corresponding to all the to-be-detected items and the number of experiment numbers contained in all the to-be-detected items, experiment numbers contained in all the to-be-detected items are distributed, and generating a second layout file suitable for the immunofixation electrophoresis apparatus. The processing efficiency of to-be-detected items can be improved, and the risk of manual operation errors is reduced.

A method for graph neural network-based recommendation is disclosed. The method comprises the following steps: receiving a user-project bipartite graph, wherein the user-project bipartite graph comprises a user node set representing a user set, a project node set representing a project set, and an edge set representing interaction between users and projects; obtaining user embedding of each user in the user set and project embedding of each project in the project set; performing an intra-layer aggregation process to generate an aggregated user embedding of each user in the user set and an aggregated item embedding of each item in the item set based on different weight coefficients of each user and each item; executing an interlayer propagation process to generate a propagation user embedding of each user in the user set and a propagation item embedding of each item in the item set based on different weight coefficients of each propagation layer; and making recommendations for the users in the user set based on the propagation user embedding and the propagation item embedding.

The invention discloses a project transaction pushing method and system based on artificial intelligence, and belongs to the technical field of semantic modeling and graph neural networks. Project information submitted by at least one project party is collected, and text description of the project information is converted into semantic vectors; historical investment records and preference information of multiple capital parties are collected, and a multi-dimensional semantic preference vector is constructed; inputting the preference vector into a pre-training graph neural network model, and performing mapping and enhancement on a transverse semantic association relationship between the items; short-term interest features are extracted in combination with the current active state, the recent behavior and the fund flow condition of the capital party, and the short-term interest features are fused with long-term preferences to generate a dynamic weight vector; performing weighted sorting on the items based on the priority score and the dynamic weight, and outputting a personalized candidate item set; according to the method, the accuracy, timeliness and individuation degree of project matching are improved, and the method has good popularization value.

The invention is applicable to the technical field of recommendation, and provides a collaborative filtering recommendation method based on double popularity constraints, which comprises the following steps of: firstly, constructing a user item score table and an item attribute relation table; secondly, obtaining normalized item popularity and normalized attribute popularity, and obtaining constraint factors by using double popularity; performing similarity calculation by using a dual popularity constraint factor and a Pearson similarity formula to obtain adjacent users of the target user; according to the scoring information of the adjacent users of the target user, performing prediction scoring on the item set which is not scored by the target user to obtain a prediction scoring table of the target user; and finally, extracting first N items with the highest score value from the prediction score table as recommendation results and providing the recommendation results to the target user. According to the method, by constraining the item popularity and the attribute popularity, the popularity deviation is relieved, and the long-tail effect is relieved.

The invention provides a page display method and device, equipment, a storage medium and a product, and relates to the technical field of computers. The method comprises the steps that an information input item set is displayed in a first area in a preset page, preset media content is displayed in a second area in the preset page, and the preset media content comprises image information related to at least one information input item. According to the technical scheme, the information input item set and the image information related to the at least one information input item are displayed in the same page at the same time, the related image information of the information input item can be displayed more visually and vividly, a user is helped to know the information input item more comprehensively, and the user experience is improved. And the information input efficiency, accuracy and success rate are improved.

The invention discloses a data poisoning attack method based on a deep learning recommendation algorithm, and the method comprises the steps: precisely screening abnormal scoring users and interaction items thereof through a statistical method in combination with a non-consistency measurement NCM and P-value, and constructing a selection item set; secondly, injecting a forged user, endowing a selected item set and a target item set with the highest score, scoring a filling item set according to normal distribution, and forming a user-item score matrix R'with poisoning data; a double-layer iterative optimization framework is adopted, and an outer layer maximizes a target project hit rate through attack loss CWLoss, enhances loss SFALoss through spectral features and hits attack traces through regularization. And the inner layer uses the user-project scoring matrix R'of the poisoning data to update the model parameters based on the deep learning recommendation algorithm until the attack converges. According to the method, the recommendation result can be remarkably manipulated under the condition of only a small number of poisoning users, high concealment and high attack efficiency are both achieved, and an effective means is provided for evaluating and improving the robustness of a recommendation system.

A multi-functional collection box enhances the security of the surrounding environment and convenience of item collection. The box features one or more chutes for receiving various categories of items, a structural pole equipped with outward facing cameras for capturing and transmitting security footage, and flood lights for illumination. An electronic interface provides users with receipts based on the weight of their donations, measured by an internal scale within the box.

The invention discloses a to-do list collection and processing method and system based on Java, and belongs to the technical field of computer software, collection and processing of to-do lists are achieved through feature dictionary matching, and the method comprises the steps that multi-source data are collected, a to-do feature dictionary is constructed, and to-do information is recognized through a multi-dimensional matching algorithm; the rear end constructs a layered architecture based on the SpringBoot and the SpringMVC, and the SpringMVC is used for providing a standard RESTfulAPI (Application Program Interface) and supporting to-do query and state updating functions; the front end adopts Vue3 + ElementPlus to realize a response type Web interface, and a back-end interface is called through Axios to realize efficient collaboration of the front end and the back end. According to the method, automatic access of multi-source heterogeneous data and extraction of the information to be handled are realized, the accuracy of identification of the backlogs is improved, and automatic classification, priority ranking, reminding triggering and state updating of the backlogs are realized.

Systems and methods are described for disambiguating user input based on a physical location of items in a vicinity of a user. The system determines that a query received from a user contains an ambiguity. In response, the system identifies several items in the physical vicinity of the user. Then, the system analyzes the identified plurality of items to determine whether the plurality of items forms a first ensemble of items or a second ensemble of items. If the plurality of items forms a first ensemble of items, the system performs a search using the search query and a first keyword related to the first ensemble of items. If the plurality of items forms a second ensemble of items, the system performs a search using the search query and a second keyword related to the second ensemble of items. The system then outputs results of the performed search.

This invention belongs to the field of hydropower project cost management technology, specifically providing a method and system for reusing historical cost data of hydropower projects based on standardized coding. The method includes: constructing a material master data dictionary and assigning corresponding master data codes to each material item; collecting core business data from historical projects, preprocessing and fusing the personnel, materials, and machinery data, and establishing a mapping relationship between the personnel, materials, and machinery data and the corresponding material items in the material master data dictionary based on comprehensive confidence, generating a standardized historical database; constructing project feature vectors for new projects and historical projects based on the standardized historical database, determining the weights of each dimension through a combined weighting method, calculating the comprehensive similarity of projects, and outputting a set of similar historical projects; and reusing historical cost data based on the set of similar historical projects, transforming the historical cost data into project cost data for new projects. This invention can improve the standardization of historical cost data and the efficiency of cross-project reuse.

The invention discloses a path selection method and device, a nonvolatile storage medium and electronic equipment. The method comprises the steps that after an emergency event occurs, a plurality of candidate paths are determined, and the candidate paths are passing paths of emergency event processing resources for processing the emergency event; according to a preset evaluation item set, determining initial comprehensive evaluation values respectively corresponding to the plurality of candidate paths; determining a target historical emergency path according to the event type of the emergency event, and correcting the initial comprehensive evaluation value according to the evaluation value of the target historical emergency path to obtain a target evaluation value of the candidate path; and determining the candidate path with the highest corresponding target evaluation value as a target path. According to the method and the device, the technical problem that the optimal path cannot be determined from the candidate paths as the target path due to the fact that only a single evaluation index is considered when the target path is determined in related technologies is solved.

The invention relates to the technical field of health data processing, in particular to an intelligent physical examination item recommendation and result analysis method and platform and electronic device.The method comprises the steps that historical physical examination information and historical training data of a target user are collected, and physical characteristics of the target user are constructed based on the historical physical examination information; the method comprises the steps of establishing an environment adaptation feature set of a target user according to historical training data, collecting training data and training environment data of the target user in real time, determining a real-time physical ability state by combining the training data with physical ability features of the target user, and updating the real-time physical ability state of the target user according to the environment adaptation feature set of the target user. And extracting a post feature item set according to the post attribute of the target user, and recommending physical examination items to the target user according to the real-time physical fitness state of the target user in the monitoring period and the post feature item set. According to the invention, the recommendation efficiency of the physical examination items is effectively improved.

The application discloses a virtual item collection method and device, an electronic device and a storage medium. The method comprises the following steps: in response to a collection operation for a virtual item, displaying an item collection page, wherein the item collection page comprises a virtual object in a first state; in response to an interaction operation triggered for the virtual item, causing the virtual object to change from the first state to a second state; when it is determined that the interaction operation meets a first preset condition and it is determined that the second state of the virtual object meets a second preset condition, triggering collection of at least part of the virtual item. The scheme can improve the interaction of a user sharing a virtual item and improve the interest of virtual item collection.

Partner-based item recommendations are described. A computing system receives historical data associated with user engagement and transactions for a set of items. The computing system selects a set of candidate items from the set of items based on a set of categories associated with the set of items, a set of popularity metrics associated with the set of items, and a set of similarity metrics associated with the set of candidate items and the set of items. The computing system generates, for each candidate item, a respective user preference score based on the historical data and using a neural network including an attention mechanism and a weighted loss function. The computing system orders the set of candidate items according to the user preference scores and a category distribution for the set of items. The computing system outputs at least one candidate item based on the ordered set of candidate items.

Methods and systems are disclosed for sharing collections of content items in chat sessions. The methods and systems receive a request to share a first content item and present a GUI comprising a first set of options and a second set of options, the first set of options being associated with adding the first content item to a collection of content items that is accessible to a plurality of recipients, the second set of options being associated with sending the first content item to individual recipients. The methods and systems determine a set of target recipients of the first content item and select a content sharing link between a first link to the collection of content items and a second link directly to the first content item. The methods and systems send, to a target recipient, the content sharing link that has been selected.

Methods, systems, and computer programs are presented for proactively generating content based on performance of previous content. One method includes an operation for transmitting a set of first items for presentation. The first items are multimodal, and each first item has values for attributes associated with the first item. The method further includes tracking performance of the transmitted set of first items, selecting values of attributes based on the tracked performance, and proactively generating, using one or more generative artificial intelligence (GAI) tools, a set of second items based on the selected values of the attributes. Further, the method includes operations for providing a user interface (UI) with an option to select from the set of second items, receiving in the UI a selection of selected second items for transmittal, and transmitting the selected second items to one or more users.

Systems and methods for receiving a set of media content items including a geohash defining a captured time and a captured location of the media content item, identifying a first subset of media content items from the set of media content items comprising a geohash that equals a precision level threshold, and identifying a second subset of media content items from the set of media content items that include a geohash that exceeds the precision level threshold. The system also includes generating a timeline media content item collection including the second subset of media content items each including a geohash that exceeds the precisions level threshold, and causing display of a media content collection interface, the media content collection interface including the timeline media content item collection.