43 results about "Patent retrieval" patented technology

The invention discloses a patent evaluation method based on a feature-efficacy matrix and a large language model, which performs feature-efficacy matrix analysis and construction on a patent text set to be processed by using text generalization and downstream task potential of the large language model, and does not need to adopt manual annotation, thereby reducing the cost. Wherein at least one feature-efficacy matrix and a patent summary contained in a patent text to be processed are obtained in the analysis and construction step of the large language model. According to the method, a vector and efficacy interval mixed retrieval mode is constructed based on the feature-efficacy matrix, and complex query requirements of patents are accurately matched. Meanwhile, a multi-dimensional patent quality evaluation mode based on a feature-efficacy matrix is a comprehensive evaluation mode from the dimensions of novelty, efficacy, popularity and the like and is applied to patent retrieval, and the analysis effect of related patents is further improved.

The invention belongs to the technical field of information retrieval, and discloses a multi-modal patent semantic retrieval method and system.The reliable cross-modal relation is recognized by establishing relation anchor points, extracting local cross-modal relation representation around the anchor points and verifying the cross-modal consistency of the relations, and further, the multi-modal patent semantic retrieval method and system have the advantages that the reliability of the multi-modal patent semantic retrieval is improved; technical features / components and successfully aligned cross-modal relationships are constructed into graph structure representation, and similarity calculation is performed by using a graph matching algorithm, so that the overall logic and details of technical schemes can be more effectively understood, schemes only containing the same technical features but having different relationship structures can be distinguished, and the accuracy of the technical schemes can be improved. The accuracy and correlation of multi-modal patent retrieval can be improved.

The specification discloses a patent retrieval method, device and equipment for multiple retrieval scenarios. Patent field language models are obtained by training general language models with fields of patent documents as training data. For multiple retrieval scenarios, positive and negative samples of each patent document are screened through different weight multi-path recall and random sampling, and training data of a patent semantic representation model of each retrieval scenario is constructed to train the patent field language models, and then the patent semantic representation model of each retrieval scenario is obtained to generate semantic vectors of each field of the patent, and the semantic vectors are stored in a patent retrieval vector database of each retrieval scenario. According to the semantic vectors of the retrieval fields, similar patent documents are searched from the patent retrieval vector database of the target retrieval scenario. Through the above scheme, the semantic representation ability of the patent semantic representation model for the patent fields of different retrieval scenarios is improved, and the retrieval precision is improved.

This invention discloses a patent matching method and apparatus based on user needs. The method includes: inputting historical patent texts into a patent feature extraction model to generate multiple module-fused feature vector matrices for each historical patent text; vectorizing all module-fused feature vector matrices and storing them in a patent feature library to be matched; after obtaining user needs, extracting a set of feature words corresponding to the user needs; dynamically adjusting the parameters of the BM25 algorithm based on the technical field and word frequency distribution corresponding to the user needs; and then filtering patent texts from the patent feature library to be matched based on the set of feature words to obtain a coarsely ranked set of patent texts; generating a module-fused feature vector matrix corresponding to the user needs based on the patent feature extraction model; and filtering patent texts from the coarsely ranked set of patent texts based on the module-fused feature vector matrix corresponding to the user needs. This invention can improve retrieval efficiency and the matching degree between patent retrieval results and user needs.

The application relates to a patent retrieval method and device based on big data, equipment and a storage medium, and relates to the technical field of computers.The method comprises the following steps: obtaining and expanding a keyword input by a user to obtain an expanded keyword set, determining a preliminary retrieval result set according to the keyword set, determining a relevance score matrix according to a text vector representation of each patent in the preliminary retrieval result set, determining a high-relevance patent set according to the relevance score matrix and the application date time sequence of each patent in the high-relevance patent set, determining a technology trend prediction result of the high-relevance patent set according to the high-relevance patent set, the corresponding application date time sequence and the relevance score matrix, and generating a retrieval result set for the user according to the high-relevance patent set, the corresponding technology trend prediction result and the relevance score matrix.The application can improve the accuracy and comprehensiveness of retrieval.

The application discloses a patent retrieval method based on a feature-effect matrix and a large language model, and feature-effect matrix analysis and construction are performed on a patent text set to be processed by using the text generalization and downstream task potential of the large language model, without manual marking, so that the cost is reduced. In the analysis and construction step of the large language model, at least one feature-effect matrix contained in the patent text to be processed and a patent summary are obtained. The application constructs a vector, an effect interval mixed retrieval mode based on the feature-effect matrix, accurately matches the complex query requirements of the patent, and further improves the retrieval effect of the related patent.

The invention discloses a patent retrieval method based on a feature-efficacy matrix and a large language model, which performs feature-efficacy matrix analysis and construction on a patent text set to be processed by using text generalization and downstream task potential of the large language model, and does not need to adopt manual annotation, so that the cost is reduced. Wherein at least one feature-efficacy matrix and a patent summary contained in a patent text to be processed are obtained in the analysis and construction step of the large language model. According to the method, a vector and efficacy interval mixed retrieval mode is constructed based on the feature-efficacy matrix, complex query requirements of patents are accurately matched, and the retrieval effect of related patents is further improved.

The present invention provides a method and system for training a machine learning-based patent search or novelty assessment system. The method includes providing a plurality of patent documents, each of which has a computer-readable claim block and a full specification block, wherein the full specification block includes at least a portion of the specification of the patent document. The method also includes providing a machine learning model and training the machine learning model using a training data set, wherein the training data set includes data from the patent documents used to form the trained machine learning model. According to the present invention, the training includes using claim block and full specification block pairs from the same patent document as training cases in the training data set.

The invention relates to the technical field of database retrieval, in particular to a patent retrieval type intelligent construction method and device and a storage medium. The method provided by the invention comprises the following steps: acquiring a natural language of a patent queried by a user; inputting the natural language into a large language model for intention understanding to obtain intention information; the intention information at least comprises a retrieval technology intention, a time range intention, a region intention and an application subject intention; converting the intention information into an sql query statement; executing the sql query statement in a pre-established data table to obtain key information; generating structured information according to the key information; inputting the structured information and the mode information of the patent database into a cue word template, and inputting the cue word template into a large language model to obtain a retrieval formula output by the large language model; and performing retrieval in a patent database based on the retrieval formula. According to the method, an accurate retrieval formula is constructed by utilizing a large language model.

The invention provides a design method and device of a patent retrieval statement, a medium and equipment. The design method comprises the following steps: identifying a logical operation symbol in a retrieval formula; based on the logic operation symbols, splitting the retrieval formula into at least one retrieval word group; wherein the search word group comprises at least one search word; identifying the category of the search word group; based on the category of the search word group, generating an extended search word of the search word group; based on the category of the search word group, the search word group and the extended search word, generating a search type search statement; according to the method, a retrieval formula is split into retrieval word groups according to logical operation symbols, extended retrieval words are generated by identifying the categories of the retrieval word groups, and more accurate retrieval statements are generated by combining the categories of the retrieval word groups, the retrieval word groups and the extended retrieval words, so that patent documents more conforming to expectations are provided for retrieval personnel, and the retrieval efficiency is improved. The retrieval difficulty is reduced; and the retrieval efficiency and effect are improved.

This invention relates to a method for patent retrieval and technology monitoring of multi-source heterogeneous data. It constructs a multimodal semantic representation including text, images, and time-series fragments, and generates a joint temporal representation vector by combining attention perturbation entropy and term evolution distance. A unified semantic space is established based on historical technology anchors, dynamically monitoring and quantifying semantic shifts to achieve real-time determination of drift trends. When substantial semantic drift occurs, local semantic correction is performed using gated contrastive learning and an anchor-aware adaptation layer to guide the orderly alignment of newly added modal features, ensuring the continuation and expansion of the original semantic topology. Simultaneously, an anchor decay and candidate new anchor generation mechanism is designed, combining cross-validation from patent citations, industry reports, and policy documents to achieve incremental expansion of the semantically stable anchor set. This scheme effectively improves the fusion accuracy, real-time adaptability, and dynamic topology update capability of multimodal patent data.

The invention discloses an intellectual property information analysis and management method based on multi-dimensional data, and the method comprises the following steps: obtaining patent original data, constructing a patent record, and carrying out the preprocessing of an intellectual property data set; dividing the patent text data into a plurality of coding units according to a preset field range, and constructing vectorization representation for each patent; generating an adjusted sub-query set for the retrieval input based on a query decomposition algorithm; obtaining a sub-query mark sequence and a comprehensive candidate result set; and calculating a comprehensive sorting score, and generating a retrieval result list. According to the method, by introducing a patent retrieval framework combining multi-dimensional query decomposition and ConstBERT fixed multi-vector document representation, executable sub-query configuration can be stably generated and fine-grained semantic matching can be realized in a scene in which a long retrieval type, multiple limits and exclusion conditions coexist; and the accuracy, controllability, traceability and large-scale application performance of patent retrieval are improved.

The application provides an optimization method and device for improving cross-domain retrieval effect, and belongs to the technical field of data processing systems under the internet industry, such as a system based on a specific computing model or a data processing system or method technology field specially used for business. The feature coding model provided by the application can cross the difference in the visual dimension between e-commerce commodity illustrations and appearance patent illustrations. The system first extracts feature vectors of text and images by using natural language processing technology, computer vision technology and a deep learning model, then realizes cross-modal mapping of images and text descriptions and cross-domain mapping between e-commerce commodity illustrations and appearance patent illustrations by using a contrast learning method, so that the similarity between different image forms can be captured, common features between e-commerce commodity illustrations and appearance patent illustrations can be effectively identified, and the accuracy of cross-domain appearance patent retrieval of an appearance retrieval recall system can be improved.

The application discloses a kind of patent retrieval statistical engine based on Elasticsearch, comprising: standard request parameter acquisition module, for obtaining standard request parameter;Elasticsearch query statement conversion module, for converting standard request parameter into Elasticsearch query statement;Patent retrieval module, for carrying out patent retrieval in Elasticsearch using Elasticsearch query statement;Standard response result obtains module, for converting Elasticsearch retrieval result into standard response result output;The application defines an interface specification of patent retrieval statistics, implements patent retrieval statistical engine based on Elasticsearch, can handle a variety of patent retrieval statistical needs including complex condition retrieval, multidimensional data statistical analysis, patent deduplication document merging and asynchronous patent retrieval.

The invention discloses an intelligent patent retrieval system and method supporting multi-dimensional features, relates to the technical field of patent information retrieval analysis, and is used for solving the problem of inaccurate complex retrieval. Comprising an access module, a slicing module, an establishment module, a planning module, a fusion and arrangement module and an evidence module, the access module unifies names, abstracts, claims, specifications and classifications, subjects, time, families, citations and legal states and generates literature keys and relation keys, and the slicing module unfolds a reference chain to analyze a limiting structure, extracts components, relations, actions and constraint fragments and records positions. The establishment module constructs an inverted table, a vector table, a field table and a relation table, the planning module generates an execution sequence, an expansion rule and a rollback rule, the fusion module recall multiple tables and performs fusion sorting with a coverage matrix, and the evidence module aligns fragments to generate evidence fingerprints, so that the system can continuously correct synonymous expressions, slice boundaries and execution strategies, noise hit is reduced, and the system performance is improved. And the continuous adaptation capability of the retrieval result is improved.

The invention provides a method and a device for processing patent drawings based on a Transform model, and relates to the technical field of patent retrieval, the method is applied to a server, and the method comprises the following steps: obtaining patent drawings with drawings marks; carrying out splitting and linear mapping processing on the patent attached drawings to obtain patent attached drawing layers; the patent attached drawing layer is input to a Transform model, a feature pyramid is obtained, the feature pyramid comprises semantic features and spatial features marked by attached drawings, the semantic features are text content marked by the attached drawings, and the spatial features are the positions of the attached drawings marked in the patent attached drawings; the feature pyramid is input into an FPEM stacking model, an enhanced feature map is output, and the enhanced feature map comprises semantic features and spatial features marked by attached drawings; and feature recognition is performed on the enhanced feature map, feature processing is performed on the enhanced feature map, the text position of the attached map mark is obtained, and the feature of the attached map mark can be accurately extracted.

This application belongs to the field of information retrieval technology and discloses a multimodal patent semantic retrieval method and system. By establishing relation anchor points, extracting local cross-modal relation representations around the anchor points, and verifying the cross-modal consistency of these relations, reliable cross-modal relations can be identified. Furthermore, technical features / components and successfully aligned cross-modal relations are constructed into a graph structure representation, and a graph matching algorithm is used to calculate similarity. As a result, the overall logic and details of the technical solution can be understood more effectively, and solutions containing only the same technical features but with different relational structures can be distinguished, thereby improving the accuracy and relevance of multimodal patent retrieval.

To solve a problem of preparing a large amount of learning data by utilizing knowledge of a skilled patent engineer when learning a specification text of a patent publication and creating a language model dedicated to patent publications for evaluating a degree of coincidence and a degree of similarity between patent publications.SOLUTION: A patent investigation system 10 uses a sentence encoder 1 based on a language model for calculating a similarity between a text for each claim of a patent claim range of a patent publication 80 of interest and a text for each claim of patent claims of an investigation target patent publication 90. The sentence encoder 1 deletes stop words from texts of patent claims and specifications of a large number of patent publications having the same section symbol of IPC classification described at the head of the patent publication 80 of interest by a division unit 20, divides the texts into segments divided at each punctuation mark equal to or less than one fourth of the number of tokens of an input limitation of the language model, and performs relearning using the segments.SELECTED DRAWING: Figure 4

1. Name of the product in this design: Intelligent Retrieval Graphical User Interface for Electronic Devices. 2. Intended use of this design: for use in an electronic device. 3. The key design features of this product are the graphical user interface displayed in the electronic device. 4. The image or photograph that best illustrates the design's key features: the front view. 5. Purpose of the graphical user interface: This graphical user interface is used for novelty searches and generating corresponding search reports, such as patent searches or Freedom to Operate (FTO) searches. The main view is the initial interface, displaying a "Start Agent" control. Interface state diagram 1 shows the interface state obtained after clicking the "Start Agent" control in the main view, displaying a "Confirm Technical Features, Start Search" control. Multiple technical features listed on this interface can be selected or deselected using the corresponding checkboxes. Interface state diagram 2 shows the interface state obtained after clicking the "Confirm Technical Features, Start Search" control in interface state diagram 1. Interface state diagram 3 shows the interface state automatically presented after the technical features in interface state diagram 2 are identified. Interface state diagram 4 shows the interface state obtained after swiping up on interface state diagram 3. The interface states obtained are shown in Figure 5; Figure 5 shows the interface state obtained after swiping up on Figure 4, which displays the "Create Report" control; Figure 6 shows the interface state obtained after clicking the "Create Report" control in Figure 5; Figure 7 shows the interface state obtained after swiping up on Figure 6; Figure 8 shows the interface state obtained after swiping up on Figure 7; Figure 9 shows the interface state obtained after swiping up on Figure 8; Figure 10 shows the interface state obtained after swiping up on Figure 9; Figure 11 shows the interface state obtained after swiping up on Figure 10.

The application provides a patent drawing processing method and device and electronic equipment, and relates to the technical field of patent retrieval. The method is applied to a server and includes: acquiring a patent drawing with a drawing mark; splitting and linearly mapping the patent drawing to obtain a patent drawing layer; inputting the patent drawing layer into a Transformer model to obtain a feature pyramid, wherein the feature pyramid includes semantic features and spatial features of the drawing mark, the semantic features are text content of the drawing mark, and the spatial features are positions of the drawing mark in the patent drawing; inputting the feature pyramid into an FPEM stacked model to output an enhanced feature map, wherein the enhanced feature map includes the semantic features and the spatial features of the drawing mark; performing feature recognition on the enhanced feature map, performing feature processing on the enhanced feature map, and obtaining a text position of the drawing mark. The application has the effect of accurately extracting features of the drawing mark.

The invention provides an optimization method and device for improving a cross-domain retrieval effect, and belongs to the technical field of data processing systems in the Internet industry, such as systems based on a specific calculation model or data processing systems or methods special for business. The feature coding model provided by the invention can cross the difference between an e-commerce commodity legend and an appearance patent legend in a visual dimension, and the system firstly uses a natural language processing technology, a computer vision technology and a deep learning model to extract feature vectors of a text and an image; and then cross-modal mapping of image and text description and cross-domain mapping between the e-commerce commodity legend and the appearance patent legend are realized by a comparative learning method, so that similarity between different image forms can be captured, common features between the e-commerce commodity legend and the appearance patent legend can be effectively identified, and the accuracy of the e-commerce commodity legend and the appearance patent legend can be improved. And the cross-domain appearance patent retrieval accuracy of the appearance retrieval recall system can be improved.

The invention belongs to the field of informatization construction, and particularly relates to an intelligence information database construction system and method for function decoupling, and the system comprises a patent collection unit, a patent storage unit, an information verification unit, a patent retrieval unit, and a transmit-receive unit. Decoupling of a patent thematic database and other units of the whole process management and control system is achieved, the adopted mode is a data request-data release mode, a data request node and a data release node do not have direct association, data storage and receiving and sending can be directly carried out through the association of address information, and the data release efficiency is improved. And personnel information of different units and different departments does not need to be identified, so that the patent access efficiency is effectively improved. The method can better improve the access quality of the patent thematic database at present with more and more personnel and more and more patent data, has a great application prospect, and avoids the complexity of access among different interfaces at the same time.

An appearance patent picture retrieval method and system, electronic equipment and storage medium are provided, and relate to the field of model training. The method comprises the following steps: constructing a test set based on a plurality of commodity pictures and patent picture labels corresponding to each commodity picture; evaluating the retrieval performance of a plurality of preset candidate multi-modal models based on the test set, and taking the candidate multi-modal model with the best retrieval performance as a base model; performing first-stage incremental pre-training on the base model based on a first training set to obtain a pre-trained model, performing second-stage contrastive learning training on the pre-trained model based on a second training set to obtain a target retrieval model, receiving a target commodity picture to be retrieved, performing feature extraction on the target commodity picture based on the target retrieval model, and outputting a corresponding patent picture list. The technical solutions provided in the present application can improve the accuracy of appearance patent retrieval results.

The present invention relates to an intelligent recommendation system for patent conversion and application objects, which belongs to the field of intelligent recommendation technology and solves the problem of low accuracy of existing patent conversion and application object recommendation methods. The system includes: a multi-dimensional technical summary module, which uses a large model to obtain technical summaries of patents to be converted in different dimensions, and obtains technical summary results of patents to be converted in corresponding dimensions after reviewing the technical summaries; a multi-dimensional candidate patent retrieval module, which extracts features from the technical summary results of patents to be converted in different dimensions, and performs intelligent retrieval based on the extracted technical features to obtain candidate patent sets of patents to be converted in corresponding dimensions; a patent conversion and application object intelligent recommendation module, which obtains the co-occurrence degree, average similarity and operation degree of co-occurring patents between candidate patent sets of different dimensions of patents to be converted to determine the recommendation degree of each co-occurring patent; and the co-occurring patent with the largest recommendation degree is used as the patent conversion and application object.

The invention relates to the technical field of bridge construction, in particular to a bridge hanging basket pre-pressing construction method which comprises a construction process and an intelligent monitoring system. The intelligent monitoring system is combined with pre-pressing construction for construction; the construction process comprises the following steps that 1, a pre-pressing model is constructed; step 2, graded loading control is carried out, and four-stage loading is carried out; step 3, deformation cooperative monitoring; 4, intelligent early warning and adjustment are carried out; 5, data visualization feedback is carried out; the counterweight device is modularized; the innovation point is that the modularized counterweight units which can be magnetically connected are adopted, compared with a traditional stacking mode, the installation efficiency is improved by three times, and according to actual measurement data, the installation time of a single module is shorter than or equal to 90 s; meanwhile, it is confirmed that the connection mode is not applied in advance in the pre-pressing field through patent retrieval; according to the scheme, the intelligent monitoring method is embodied in a distributed optical fiber and three-dimensional scanning cooperative monitoring mechanism, and the defect that space deformation cannot be captured through traditional single-point monitoring is overcome.

The invention provides an approximate patent retrieval method and device, electronic equipment and a storage medium, related technical keywords are extracted from to-be-retrieved patents to serve as retrieval keywords, approximate patents are retrieved in a patent database according to the retrieval keywords and retrieval strategies, and the retrieved approximate patents are recommended to a user after being sorted according to the approximation degree. According to the patent retrieval method and device, related technical features are screened from the to-be-retrieved patents for retrieval in an artificial intelligence mode according to the accurate retrieval requirement of the single to-be-retrieved patent file, the patent retrieval difficulty is lowered, the patent retrieval efficiency is improved, and the retrieval accuracy is improved.

1. Name of the product in this design: Graphical User Interface for Patent Intelligent Search of Electronic Devices. 2. Purpose of this design: An electronic device. 3. The key design feature of this product is its graphical user interface. 4. The picture or photo that best illustrates the key design points: Design 1 front view. 5. Design 1 is designated as the basic design. 6. Purpose of the graphical user interface: To optimize the search method for patent technical feature schemes. 7. Human-computer interaction method of graphical user interface: The main view of Design 1 is the graphical user interface presented when entering the search module. Clicking the "Heimei AI" button in the search bar of Design 1's main view will present the interface change state diagram 1. Clicking the "Search-based Optimization" button in the interface change state diagram 1 will present the interface change state diagram 2. Entering the technical features to be searched in the interface change state diagram 2 will present the interface change state diagram 3. Clicking the "blue arrow" button in the search-based optimization search box in the interface change state diagram 3 will present the interface change state diagram 4. Clicking the button in the interface change state diagram 1 will present the interface change state diagram 4. The "Application Search" button in the interface change diagram 4 presents the interface change diagram 5 of Design 1; the main view of Design 2 is the graphical user interface presented when entering the search module. When the technical features to be searched are directly entered in the search bar, the interface change diagram 1 of Design 2 is presented. In the state of the interface change diagram 1 of Design 2, "double-clicking the Ctrl button on the keyboard" presents the interface change diagram 2 of Design 2. Clicking the "Search Optimization" button in the interface change diagram 2 of Design 2 presents the interface change diagram 3 of Design 2. Clicking the "Application Search" button in the interface change diagram 3 of Design 2 presents the interface change diagram 4 of Design 2.

The invention discloses an associated patent determination method and device, equipment, a medium and a product, and the method comprises the steps: determining a patent retrieval mode corresponding to a currently browsed patent when the currently browsed patent is detected; wherein the patent retrieval mode is semantic retrieval or non-semantic retrieval; and if the patent retrieval mode is semantic retrieval, determining a target associated patent corresponding to the currently browsed patent from the retrieval database based on the target content in the currently browsed patent and the retrieval text corresponding to the currently browsed patent. The problem that in the prior art, in the patent retrieval process, the retrieval result is not accurate only based on keyword matching or simple semantic analysis is solved, the patent better conforming to the retrieval intention of the user is rapidly and accurately determined according to the currently browsed patent of the user, and therefore the patent retrieval experience effect of the user is improved.

The invention provides a technical blank space calibration and layout subject identification method based on suspected missed labeling patents, belongs to the technical field of patents, and solves the problems of technical blank misjudgment and the like caused by term diversity and indexing limitation in an existing patent analysis method. The method comprises the following steps: constructing a patent retrieval expression according to target technology description, performing retrieval, and constructing a large-field patent set; extracting high-purity seed samples from the large-field patent set, and constructing a high-purity seed set; modeling is carried out on the high-purity seed set from different technical perspectives, and a multi-dimensional behavior portrait is constructed; evaluating the to-be-evaluated patent based on the multi-dimensional behavior portrait, and generating a suspected label missing patent set; on the basis of the large-field patent set and the suspected missed labeling patent set, the technical blank area is calibrated, and a true blank and a false blank are distinguished; and identifying an underestimated layout main body and an emerging technical direction based on the large-field patent set and the suspected missed labeling patent set.