647 results about "Text retrieval" patented technology

A Mixed Media Reality (MMR) system and associated techniques are disclosed. The MMR system provides mechanisms for forming a mixed media document that includes media of at least two types (e.g., printed paper as a first medium and digital content and / or web link as a second medium). In one particular embodiment, the MMR system includes an action processor and method, and MMR documents with an associated action. The MMR document structure is particularly advantageous because the ability to specify different actions for different MMR documents, combined with the ability to create any number of MMR documents for a particular location on any media, allows the MMR architecture to serve as a universal trigger or initiator for additional processing. In other words, addition processing or actions can be triggered or initiated based on MMR recognition. The action processor receives the output of the MMR recognition process which yields an MMR document including at least one action. The action processor executes that action which includes various commands to the MMR system or other systems coupled to the MMR system. The MMR system architecture is advantageous because an action can be executed by pointing the capture device at a block of text, and the action is performed. Example actions include retrieving the text in electronic form to the capture device, retrieving the specification for the action, inserting data to a MMR document, transferring data between documents, purchasing items, authoring actions or reviewing historical information about actions. The MMR system includes a variety of user applications (one or more actions) initiated by the MMR recognition of a text patch such as information retrieval for a travel guide book, stock listings or advertisements; information capture such as recording content from a conference, recording and storing multimedia associated with the document, capturing information for a calendar and on the fly authoring; purchasing media files for storage on any part of an MMR document.

An information retrieval apparatus includes contents, an index data generating unit, a character frequency management data generating unit, a compressing / encrypting unit, a retrieval initializing unit, a full text retrieving unit, and a retrieval result displaying unit. The character frequency management data generating unit generates character frequency management data based on the contents. The compressing / encrypting unit compresses the contents and encrypts the character frequency management data. The retrieval initializing unit decrypts encrypted character frequency management data. The full text retrieving unit executes full text retrieval for compressed contents using the character frequency management data and index data when receiving a retrieval keyword. The retrieval result displaying unit decompresses a retrieval candidate selected from retrieval candidates and displays as a retrieval result.

The invention belongs to the information retrieval technology and provides a full-text retrieval system based on semantic analysis of relevant words. The full-text retrieval system based on semantic analysis of the relevant words comprises an inquiry information receiving module, a concept semantic analysis module based on the relevant words, a semantic knowledge base module, a retrieval module, an index database, an index module, a theme semantic analysis module based on the relevant words, a result set processing module and a data server. The full-text retrieval system based on semantic analysis of the relevant words is based on the improvement on a traditional Internet search engine, and by the adoption of the system, concept semantic analysis based on the relevant words and theme semantic analysis based on the relevant words of a document can be achieved, and users can obtain search results which are more accurate, more comprehensive and more intelligent.

The invention discloses an image retrieval method and an image retrieval system. The method includes for given query texts and / or query images, acquiring multiple similarity ordered lists of in-base images according to text relevance and image content relevance, and then returning a comprehensive ordered list by combining the acquired ordered lists and comprehensively considering the text similarity and the image content similarity. Through the multi-mode mixed retrieval mechanism, shortcomings of conventional single-mode retrieval mechanisms are overcome, respective advantages of a text retrieval method and an image content retrieval method are developed, and accuracy of image retrieval is greatly improved. Since only ordering results of single retrieval models are fused, the single retrieval models can be increased, decreased and replaced conveniently, text and image content feature retrieval models are configured flexibly, and performance of the image retrieval system is improved.

The invention discloses a cryptogram-based safe full-text indexing and retrieval system. In the system, a cryptogram index library comprises a cryptogram entry reverse index and an internal document object set; a cryptogram document library is responsible for storing and managing an encrypted XML document; a word segmentation encryption server carries out Chinese word segmentation on a plaintext document and encrypts the plaintext document item by item; a cryptogram full-text indexing server standardizes an original plaintext document into an XML document, encrypts and stores the XML document in the cryptogram document library, creates a corresponding internal document object in the cryptogram index library by combining document metamessage, and creates a cryptogram reverse index for the XML document through the cryptogram entry; and a cryptogram full-text retrieval server retrieves the cryptogram index library to obtain the internal document object set through user authority information and the cryptogram entry, obtains a corresponding encrypted XML document result set from the cryptogram document library according to a pointer, decrypts the corresponding encrypted XML document result set, and returns the decrypted corresponding encrypted XML document result set to a user. The Chinese word segmentation method, the safe and high-efficiency indexing structure and the retrieval mechanism of the invention based on the special requirements of cryptogram full-text indexing can realize the cryptogram full-text indexing integrated with an access control strategy. The cryptogram-based safe full-text indexing and retrieval system has the advantages of a safe and high-efficiency indexing process, no decrypted docuterms in the indexing process, a high recall ratio and a high precision ratio in a cryptogram environment, and the like.

The invention belongs to the technical field of cross-modal retrieval, and particularly relates to an image-text retrieval system and method based on a multi-angle self-attention mechanism. The systemcomprises a deep convolutional network, a bidirectional recurrent neural network, an image, a text self-attention network, a multi-modal space mapping network and a multi-stage training module. The deep convolutional network is used for acquiring an embedding vector of an image region feature in an image embedding space. The bidirectional recurrent neural network is used for acquiring an embedding vector of a word feature in a text space, and the two vectors are respectively input to the image and the text self-attention network. The image and text self-attention network is used for acquiringan embedded representation of an image key area and an embedded representation of key words in sentences. The multi-modal space mapping network is used for acquiring the embedded representation of the image text in the multi-modal space. The multi-stage training module is used for learning parameters in the network. A good result is obtained on a common data set Flickr30k and an MSCOCO, and the performance is greatly improved.

A new-generation industry knowledge full-text search method includes: 1, segmentation dictionary setup, namely setting up a segmentation dictionary, and storing dictionary information into a database; 2, full index setup, namely reading, dividing words and analyzing existing full-text documents called 'knowledge point documents' to set up an index file; 3, increment index setup, namely processing newly added documents and updating an index file on a hard disk; 4, memory index setup and memory segmentation dictionary setup, namely reading segmentation dictionary data in a memory to set up a memory segmentation dictionary data structure; 5, full-text search, user question standardization, word dividing, semantic comprehension, semantic extension, candidate document acquisition and candidate document ordering, wherein segmentation dictionary setup is conducted during system initialization; full index setup includes reading all knowledge documents to fully set up a hard disk index file called 'index file' for short; increment index setup is conducted when full-text files are newly added; the three events are independent from a full-text retrieval module and in independent operation.