110results about How to "Improve query accuracy" patented technology

The invention discloses an image retrieval system integrating classification with hash partitioning, which comprises a downloading module, a classification module training module, an image classification module, a characteristic extraction module, a recording table building module, a partitioning module, a request processing module, a retrieval module, a similarity acquiring module and a result returning module. The downloading module is used for downloading images so as to build an image library, the classification model training module classifies the images in the image library according tothe shapes and selects representative sample images from the image library to form a sample library firstly, then extracts characteristic descriptors of the classification bottom layer of all images in the sample library and performs trainings on the characteristic descriptors of the classification bottom layer by a support vector machine so as to obtain discriminant of each classification. Classification models are formed according to the discriminants of all the classifications. Precision ratio of the image retrieval system is increased, the problem of low recall ratio during classificationmistakes is overcome, and the retrieval speed of the image retrieval system is increased integrally.

The invention belongs to the technical field of service computation, and relates to a service and tag recommendation method based on random walk. The method comprises a service recommendation step and a tag recommendation step, wherein the service recommendation step comprises the following process of: collecting the information of a Mashup service and an API (Application Program Interface) service, abstracting the relation between the Mashup service and the API service by association, sorting each tag after performing random walk on a co-occurrence network between the tags, and recommending the Mashup service through the association relation between the tag and the Mashup service and the Mashup semantic expansion method. The tag recommendation method comprises the following process of: searching the nearest neighbor according to the similarity of the API service description texts, and performing tag recommendation of the API service through the random walk and tag sorting method. The recommendation method can be used for improving the accuracy of the service searching result, and solving the problem of tag randomness.

The invention provides a webpage retrieval method sensitive to geographical positions, a search engine method and a search engine system. Firstly, a cloud server calculates geographical relevance of selected geographical hot spots of web pages in an off-line state and calculates scores of importance of each geographical hot spot of each web page by being combined with a network link structure obtained by a grid crawling unit, the scores are recorded in meta data of each corresponding web page as fields, and the meta data of each web page are stored in a space database of the server; when a user inquires on line, the server analyzes a geographical range of an inquire statement through natural language processing, calculates the geographical relevance of the inquire statement relative to the geographical hot spots according to the distance between the inquire statement and the geographical hot spots, calls the scores of corresponding geographical hot spots of the web pages in the space database, calculates scores of the web pages in a specific inquiry on line, sequences the results in a descending order, and outputs a retrieval result at the user side.

The invention discloses a method and a system of supervised web service finding based on attribution forecast and error correction of quality of service (QoS). The method and the system of the supervised web service finding based on the attribution forecast and the error correction of the QoS are mainly used for solving the problem of accurately effectively finding a credible web service meeting requirements of a user in an opened network environment, and therefore providing important technology support for the construction of a highly-credible service application system. On the one hand, the method and the system of the supervised web service finding based on the attribution forecast and the error correction of the QoS adopt a matrix compensation technology to carry out forecast or correction of the attribution of the QoS of the deficient or anamorphic services to enable the credible services based on the attribution of the QoS to be effectively implemented, on the other hand, prior class information of the services is introduced into the characteristic extraction stage indicated by the services to enhance distinguishing capability and query matching accuracy which are indicated by the services. Indicated by an evaluation result of a real web service data set, the method and the system of the supervised web service finding based on the attribution forecast and error correction of the QoS are not only high in efficiency and accuracy of service finding, but also simple to achieve, capable of enabling a system to be flexible, low in cost, and easy to popularize.

The invention relates to a text data processing method and device and a computer readable storage medium, and relates to the technical field of natural language processing. The method comprises: acquiring keywords and entities contained in text data; determining classification information of the text data according to the keywords and the entities; creating text indexes of the text data accordingto the keywords, the entities and the classification information; creating sentence indexes of the sentences according to the keywords and the entities contained in the sentences in the text data; determining related text data of the target event by utilizing the text index; determining a matched sentence of the target event in the related text data by utilizing the sentence index; calculating a matching weight of a keyword matched with the target event in the matched sentence; and determining a keyword related to the target event according to the matching weight and the first threshold value.According to the technical scheme, the event query accuracy can be improved.

The invention provides a method for processing composite model data. The method comprises the steps of receiving and analyzing an initial search condition sent by a client side and an object required to be returned; determining an initial model based on the initial search condition and determining a target model based on the object required to be returned; determining an incidence relation between the target model and the initial model and a search path from the initial model to the target model; obtaining search methods of each model in the search path; obtaining initial model data by sending the initial search condition into the search method corresponding to the initial model, extracting search conditions needed by other models in the search path from previous model data and obtaining corresponding model data by respectively sending the extracted search conditions to the search methods corresponding to each model; extracting the object required to be returned from all the model data and returning the object required to be returned to the client side. The invention also provides a system for processing the composite model data, which is capable of realizing the method. According to the method and the system for processing the composite model data, provided by the invention, the occupation of a system resource during searching can be reduced.

The invention belongs to the technical field of information data processing, and discloses an encrypted data retrieval and sharing method and system, a medium, equipment and an application. The method comprises the steps: performing system initialization; generating an encryption matrix index; generating a keyword trap door by using the paillier key pair; carrying out ciphertext search by using two ciphertext comparison algorithms in a search stage; after the intelligent vehicle obtains the query result, transmitting the attribute set of the intelligent vehicle to the authoritative platform; enabling the authoritative platform to decrypt the related key ciphertext based on the attribute set; and updating keywords or data. When retrieval or sharing is carried out, data is encrypted, college retrieval can be realized, fuzzy retrieval can be carried out on keywords with spelling errors, results are sorted according to relevancy, and accurate retrieval is realized; safe retrieval and sharing of data are achieved, and fuzzy retrieval can be conducted on keywords with spelling errors; the system can achieve efficient and encrypted retrieval, and how to achieve efficient encryption sharing.

The invention discloses a method and equipment for sorting a document. The method comprises the following steps of: according to user query and an ontology base, extracting queried semantic information; according to the document, the query and the ontology base, extracting document semantic information; determining a relational semantic relevance between the document semantic information and the queried semantic information; and sorting the document on the basis of the relational semantic relevance. According to the method and the equipment provided by the invention, the document sorting accuracy can be effectively improved.

The invention discloses a query method and device of a test case and an establishment method and device of the test case. The query method of the test case comprises the following steps: reading parameters included in the title of a first test case when the first test case is opened, wherein the parameters are the ones for querying a second test case having incidence relation with the first test case; adding the parameters to an initial query condition to generate a first query condition; querying the second test case having the incidence relation with the first test case according to the first query condition to obtain a first result set. Through the invention, the problem that the correct result set cannot be queried due to the need of manually inputting the query title in the prior art is solved to further achieve an effect of improving the query accuracy.

The invention discloses a space keyword ciphertext retrieval method and system for semantic comprehension. The space keyword ciphertext retrieval system comprises a key module, a semantic informationextraction module, an encryption index construction module, a trap door generation module and a query module. By extracting the space objects and the semantic features queried by the user, the user can query the space objects which accord with the query intention of the user and are close to each other on the ciphertext. The ciphertext index construction mode adopted by the invention improves precision of ciphertext space keyword retrieval, and meets the query requirements of users for distance and text. In addition, the ciphertext query algorithm of the scheme improves query efficiency whileensuring the security and privacy of the space object data and the user retrieval information.

The invention relates to an image-text hybrid retrieval method and device, a storage medium and computer equipment, and the method comprises the steps: obtaining to-be-retrieved information inputted by a user, carrying out the word segmentation of text data if the to-be-retrieved information contains the text data and image data at the same time, and then carrying out the inverted index query operation, obtaining a first candidate set; inputting the image data into a pre-trained neural network model to obtain a feature vector corresponding to the image data, encoding the feature vector, and performing inverted index query operation to obtain a second candidate set; merging the documents in the first candidate set and the second candidate set, counting the hit times of the merged documentsin the first candidate set and/or the second candidate set, sorting the merged documents according to a statistical result, and returning a sorting result to a display page; and the maintenance cost of the system is effectively reduced while the retrieval accuracy is improved.

The invention discloses a text hash retrieval method based on deep learning. The method is characterized by comprising the following steps of: extracting a semantic code corresponding to each piece oforiginal vocabulary data in a word embedding matrix by using a bidirectional LSTM model; then, connecting a text convolutional neural network in parallel behind the bidirectional LSTM model, adding an attention mechanism, converting the output value of the second full connection layer into a corresponding hash code by using a sign function; reconstructing a category label by utilizing hash codes;and finally, searching vector data closest to the Hamming distance of the retrieval text hash code in the text library hash code, and completing the hash retrieval process of the retrieval text data.The method has the advantages that the learning ability of the hash model to the short text is high, the added attention mechanism can further improve the expression ability of the features, the classification layer reconstructs the category labels by utilizing hash codes, so that the hash model can utilize the label information more finely while learning binary codes, and the retrieval precisionis high.

The invention discloses a radio monitoring data query method based on buffer technology; the equipment comprises a general interface module, a data transmission module, a HBase database, a historical data arrangement module and a buffer module; the general interface module, the data transmission module, and the HBase database are connected in sequence; the HBase database and the buffer module are respectively connected with the historical data arrangement module. The radio monitoring data query method is fast in query speed, high in query precision, and can playback mass semi-structured or unstructured data in real time.

The invention relates to a k neighbor inquire method of interactive writing brush word, wherein the invention can realize the interactive index and search based on mantic of writing brush word, that user can adjust the index to improve the inquire accuracy. The inventive method comprises: first calculating the distance of each couple of writing brush words in the word base at certain threshold value, to generate one local distance diagram, and building the index based on B+tree of said diagram; when user provides one sample writing brush word, system based on demand searches the word similar to the word; the user based on dynamic feedback selects the word similar to the mantic of said word. Therefore, the invention can based on the feedback information of user dynamically adjust the local distance diagram, to avoid irrelevant words to keep high inquire accuracy.

A context sensing and complex semantic association based data space modeling method is provided. The invention relates to a data space modeling method. The method can overcome the following detects of the conventional methods: 1) context sensing performance is poor; 2) semantic relation expression ability is poor; and 3) semantic association inference ability is poor. The context sensing and complex semantic association based data space modeling method includes: a step 1, establishing a semi-structured graph model called as a complex semantic association network model COSAN of context sensing; a step 2, expressing an interpretative object of the context sensing according to the complex semantic association network model COSAN of context sensing; a step 3, acquiring a basic semantic association and a complex multi- semantic association of context sensing according to the interpretative object of context sensing; and a step 4, acquiring a semantic association inference rule according to the basic semantic association and the complex multi-semantic association of context sensing. The data space modeling method can be applied to the field of data space modeling.

The embodiment of the invention discloses a program interface information recommendation method and device, electronic equipment and a readable medium. A specific embodiment of the method comprises the steps of obtaining target query statement information for a target programming language; performing word segmentation processing on the target query statement information; generating a target label set; determining the similarity between the target query statement information and each piece of query statement information in a pre-constructed knowledge base; screening out query statement information of which the corresponding first similarity numerical value meets a screening condition from a pre-constructed knowledge base; determining the similarity between a reply information set corresponding to each piece of candidate query statement information in the candidate query statement information set and the target query statement information; sorting the program interface information corresponding to the candidate query statement information in the candidate query statement information set; and sending the program interface information sequence to a target interface for display. According to the embodiment, the development efficiency of the application program and the query precision of the program interface are improved.

The invention discloses a breakpoint continuous transmission method, which comprises the following steps of: when file downloading is interrupted, determining a breakpoint continuous transmission table entry through a table entry containing the file abstract of the downloading interruption file in a service list, and implementing breakpoint continuous transmission according to the breakpoint continuous transmission table entry. The invention also discloses a breakpoint continuous transmission system, which comprises a service provider, personal network equipment (PNE) and a service server. The method and the system can greatly reduce the time intervals between downloading interruption and breakpoint continuous transmission.