50 results about "Tibetan language" patented technology

The invention relates to a Tibetan language entity knowledge information extraction method, which comprises the following steps that: Tibetan and Chinese comparable language material information is extracted from Tibetan and Chinese text language material information; entity equivalence pairs are extracted from the Tibetan and Chinese comparable language material information; the Tibetan and Chinese cross-language entity relationship is extracted from the entity equivalence pairs; a Tibetan language "entity-attribute-value" triad is extracted from the Tibetan and Chinese cross-language entity relationship; and the triad is stored into a Tibetan language entity knowledge semantic resource library. The Tibetan language entity knowledge information extraction method solves the problem of Tibetan language training language material deficiency to a certain degree, promotes the knowledge sharing among different languages, and provides support for the study in the fields of Tibetan and Chinese cross-language knowledge questions, information retrieval, machine translation and the like.

The invention relates to a method of Tibetan language webpage text classification based on semanteme. The method of the Tibetan language webpage text classification based on the semanteme comprises the following steps: firstly, extracting text messages which are used for charactering a webpage from the Tibetan language webpage, then carrying out word classification processing on the text messages, and expressing words obtained by means of the word classification processing as a word vector space, secondly, mapping the words expressed through the word vector space to a concept of a semanteme space according to a preset Tibetan language classification body to obtain a semanteme space of a text to be classified, and at last adopting a classification algorithm to classify the semanteme space of the text to be classified according to a preset semanteme space of a training sample set. The method of the Tibetan language webpage text classification based on the semanteme carries out preprocessing on the webpage and adopting a KNN classification algorithm of the text similarity of a weighting semantic net to achieve real-time and high-efficiency classification of the Tibetan language webpage.

The embodiment of the invention provides a Tibetan language-based multi-modal emotion calculation method and system, and a server. The method comprises: firstly, obtaining Tibetan language data to beclassified; collecting video signals, voice signals and text information from the Tibetan language data; then, extracting high-level video features, high-level voice features and text features in a classification emotion corpus, respectively extracting the high-level video features, the high-level voice features and the text features, performing learning based on a deep learning model to obtain high-level fusion features, and finally, classifying the high-level fusion features in the classification emotion corpus based on SVM and storing the high-level fusion features in the classification emotion corpus. Therefore, the blank state of the Tibetan language in sentiment analysis can be filled. A basic corpus is provided for Tibetan multi-modal sentiment analysis. The Tibetan language data sentiment recognition method based on the three modes is beneficial to development of Tibetan language multi-mode sentiment analysis, the natural language processing capacity and the intelligent sentiment recognition capacity of the Tibetan language can be promoted, the artificial intelligence information processing capacity of the Tibetan language is improved, and in addition, the sentiment recognition rate of the Tibetan language data can be effectively increased under the condition of mutual fusion of the three modes.

The syllable division of Tibetan speech is an important part and essential link in Tibetan speech information processing, and lays the foundation for Tibetan speech recognition and speech database making; however, a software for the syllable division of Tibetan speech is still absent. Through extracting the time domain parameter short-time energy and short-time zero-crossing rate of Anduo Tibetan speech, the invention realizes Tibetan syllable division through a specific algorithm. The invention is used to carry out syllable division of 30 continuous Anduo Tibetan speech samples, and the division accuracy reaches to 30.6 percent.

The invention relates to a method for identifying Tibetan language webpage and its coded, including the steps of: giving a code of characteristic string, which is syllable node and/or selected high frequency syllable, in Tibetan language codefirstly; webpage character flow, the code of characteristic string as keyword, scanned and searched; calculating the frequency that accords with characteristic string coded character to appear by the couter; determining whether the webpage is Tibetan language webpage and the Tibetan language code is used according to the result of counter. The invention makes the best of the syllable structural feature of Tibetan language language and the statistics characteristic of Tibetan language word, and respectively applys the identification criteria for different code, accordingly Tibetan language webpage and non Tibetan language webpageu can efficiently be distinguished correctly, and Tibetan language coding used by the webpage is also able to be identified.

The invention provides a Tibetan character recognition device and method, and relates to the field of character recognition. According to the invention, by receiving a photographed Tibetan character document image sent by a high-speed photographic apparatus, and according to the size of the Tibetan character document image, the line spacing of each line region, each of the suspected Tibetan characters in the coordinates of a pre-established coordinate system and the area of the minimum rectangular frame of each suspected Tibetan character, an adaptive threshold range associated with features of the Tibetan characters is generated in real time; and if an attribute of the minimum rectangular frame outside each Tibetan character is not within the adaptive threshold range generated in real time, the Tibetan characters in the minimum rectangle frame are segmented or merged, so that the cases of erroneous recognition of the Tibetan characters are eliminated according to the special structure of the Tibetan language and the special byte shape, and the accuracy rate of Tibetan character recognition is raised.

The method creates a components including: building code table of Tibetan keywords and combined words of Tibetan keywords; building key assignments for key-presses of electronic terminal; building coincidence relation between dot matrix and code table. The method includes following steps: (1) reading key values input; (2) searching dot matrix font library and displaying character; (3) selecting and determining character needed to input. The invention realizes functions for inputting and displaying Tibetan.

The invention relates to a Tibetan language input method of an embedded device. The Tibetan language input method of the embedded device achieves Tibetan language input of the embedded device through a remote controller: (1) building a mapping file of a remote controller digital code and a Tibetan language character standard Unicode code; (2) storing the mapping file in an input method installation directory, and loading the mapping file in a memory when the input method is started to form a mapping table; (3) for a user input numeric key, finding a Unicode code corresponding to the numeric key in the mapping table according to the digital code formed by the numeric key; (4) finding a corresponding Tibetan language character from a word stock based on the Unicode according to the Unicode code, and displaying on an interface through a graphic program; and (5) inputting sound insulation characters, figures, half-figures and punctuation marks in the Tibetan language through the 0 key and the # key. The Tibetan language input method of the embedded device has very important significance in development of informatization of minority areas and improvement of social informatization level.

The invention discloses an intelligent rule multi-language type interpretation system and a creation method for the same, and is applied to the field of intelligent home. By the system, a foundation is laid for solving the problem of artificial intelligence (essentially a remote control system or local intelligence) in the field of intelligent home at present. The system is an integrated language interpretation environment formed by a rule editor, rule lexical checking, rule grammar checking, intermediate code generation, executable code generation, executable code file management, rule debugging, a virtual machine execution environment and the like. The system and the method are simple and easy to learn; language grammars are similar to those of natural languages of human beings, a user can understand and use the system and the method without any professional knowledge, the whole program can be described completely by pure Chinese (Tibetan language, Mongolian language and the like) besides operators, and is high in execution efficiency, and executable codes are very close to machine languages. By fusion of such a language system and an intelligent home cloud service system, the user can speak to the system to simply require the system to automatically finish a complex process, and according to the actual condition of the current market, such a technology has broad market prospect.

The present invention discloses a method and a system for constructing a Tibetan emotional dictionary based on Tibetan language features. The method comprises: matching the Chinese vocabulary ontology with emotion classification with a Chinese-Tibetan dictionary to obtain a Tibetan basic emotional dictionary; carrying out corpus training on the preliminarily collected Tibetan microblogging information by using the Word2vec tool to obtain a synonym set of the corpus training vocabulary, and taking the synonym set as an extended candidate word set; calculating the weight variance of each extended candidate word; and screening the extended candidate words according to the weight variance to obtain emotional extension words. According to the method for constructing the Tibetan emotional dictionary based on Tibetan language features disclosed by the present invention, the Chinese vocabulary ontology is matched with a Chinese-Tibetan dictionary to obtain the Tibetan basic emotional dictionary, corpus training and screening is carried out on the Tibetan microblogging information by using the Word2vec tool, and extension is carried out based on the Tibetan basic emotional dictionary, so that more Tibetan emotional vocabulary is provided, and emotion of the current Tibetan microblogging information expression is accurately analyzed.

The invention belongs to the technical field of Tibetan language information processing, and discloses a Tibetan language word vector representation method fusing components and character information,and the Tibetan language word vector representation method fusing the components and the character information comprises the steps: directly fusing the components and the character information into amodel TCCWEI represented by the Tibetan language word vector; fusing the component information into the vector representation of the character, and then fusing the component information and the information of the character into a model TCCWEII represented by the Tibetan word vector; and fusing components and Tibetan word vectors of character position information into the TCCWEII model to represent a model TCCWEII + P. According to the Tibetan language word vector representation model TCCWE provided by the invention, compared with the current optimal word vector representation, the Tibetan language word vector representation model TCCWE is improved by 8% on the Tibetan language similarity evaluation set TWordSim215. Compared with the current optimal word vector representation, the Tibetanlanguage correlation evaluation set TWordRel 215 is improved by 7%, and the semantic and correlation of the found similarity/correlation words are very high.

The invention provides a Tibetan language input method applied to a mobile telephone, relating to a character input method and concretely designing a mobile telephone input method applied to Tibetan language. In the invention, number keys and the combination keys of the number keys and direction keys are utilized to correspondingly display a group of Tibetan language letters, every Tibetan language letter group corresponds to one number key or one functional key, and then the corresponding number keys or the corresponding functional keys are selected to affirm the Tibetan language letters required to be input, and are combined into required Tibetan language characters and sentences according to the Tibetan language letters; and the application of the corresponding letters is simple and fast. The invention has the advantages of simple input and comprehensive input of the Tibetan language, is convenient for Tibetan to use a mobile telephone to send out Tibetan language short messages, enriches the daily life of the Tibetan and is convenient for the contact among the Tibetan.

The present invention provides a method of converting gestures into a Chinese-Tibetan bilingual speech. The method comprises the steps that sample data is used for carrying out gesture recognition ongestures to be recognized to obtain meanings of the gestures; the meanings of the gestures are expressed in a Chinese-Tibetan bilingual form to obtain a semantic definition of the gestures, and context-related annotations of the gestures are generated according to the semantic definition of the gestures; a speaker-related Tibetan model or a Mandarin model is obtained through self-adaptive trainingof the speaker and using the training corpus of a specific speaker in Mandarin or Tibetan, and the Tibetan speech or Mandarin speech are synthesized by using a speaker-related Tibetan model or a Mandarin model and the context-sensitive annotations of the gestures. The method can convert the input static gestures and dynamic gestures into Mandarin or Tibetan language, can promote the daily communication activities of speech impaired persons and normal persons, solves the speech output problem in the communication between disabled persons and the normal persons, and can further be applied to the aspects such as computer assisted instruction for deaf and dumb, bilingual broadcasting of TV programs.

The invention discloses a Tibetan language encoding and decoding method and device, and belongs to the field of communication. The encoding method includes the steps that a Tibetan language syllable is obtained; characters at all the positions are sequentially judged from left to right; if the current position includes one character, a preset code table is searched to obtain a code corresponding to the character; if the current position includes multiple characters arrayed vertically, the code table is searched for all the characters arrayed vertically sequentially in the first consonant and later vowel sequence and the top-to-bottom consonant sequence; in the encoding process, a character space mark is added between every two adjacent codes, and the Tibetan language syllable codes are obtained after compiling of the characters at all the positions in the Tibetan language syllable is completed. The scheme of achieving code encoding and decoding based on the minority language is given, which greatly facilitates Tibetan language communication, and the Tibetan language encoding and decoding method plays a great role in emergency communication and security and secrecy communication.

The invention relates to a service system for automatic generation of Tibetan reading questions in primary schools. The system comprises a Tibetan reading corpus construction model and a Tibetan reading text question generation model. According to the Tibetan reading corpus construction model, the Tibetan reading corpus is obtained by extracting feature data of primary school Tibetan articles and designing a mixed multi-strategy text screening model; The Tibetan reading text problem generation model comprises a coding end and a decoding end, and the coding end uses a bidirectional RNN network and an attention mechanism; The decoding end uses a one-way RNN network, an attention mechanism and a replication mechanism. Through the designed mixed multi-strategy text screening model, Tibetan Chinese articles suitable for being read in primary schools can be screened out from large-scale encyclopedia Tibetan Chinese texts. And an end-to-end automatic problem generation model is designed, so that the problems that the Tibetan language reading teaching materials in primary schools are few, the updating speed is low, the manual question setting amount is small and the like are solved, and the development of Tibetan language teaching in national regions is promoted.

The invention provides a Tibetan language paper plagiarism detection method and system, and relates to the technical field of modern education. Aiming at three different plagiarism phenomena of continuous text plagiarism, semantic rewriting plagiarism and translation plagiarism, the invention provides a longest common subsequence algorithm and an improved twin long-short-term memory network methodrespectively. Academic paper pre-detection based on abstract document vectors and a weight distribution strategy based on chapter positions are adopted to improve retrieval efficiency.

The invention relates to the field of artificial intelligence, the invention provides a Tibetan language speech recognition system based on an HMM-DNN (hidden Markov model-deep neural network). According to the Tibetan language speech recognition method based on the HMM and the DNN, a deep learning training model is combined with Tibetan language which is a low-resource corpus, a Tibetan language-based establishment model is trained, Tibetan language speech is recognized, and the human-computer interaction efficiency of Tibetan people is improved, and the Tibetan language speech recognition method based on the HMM and the DNN comprises the following steps of 1, recording Tibetan language speech data; 2, carrying out data preparation; 3, constructing a language model and a pronunciation dictionary; 4, training a single-phoneme model; 5, training a three-tone sub-model; 6, performing linear discriminant analysis and maximum likelihood linear transformation, and performing decoding and alignment; 7, carrying out speaker adaptive training; and 8, carrying out model training. The method is mainly applied to Tibetan language speech automatic recognition occasions.

The invention provides a Chinese and Tibetan language data translation management and evaluation method under a domestic operating system, which can perform overall management on tens of millions of Tibetan language data of a large amount of software in the domestic operating system based on Linux, so that the management cost of Chinese and Tibetan language maintenance of the domestic operating system is effectively reduced; comprising the following steps: analyzing all software source codes under a domestic operating system based on Linux to obtain source entries, and constructing a source entry data set; in response to update of the translation data of the source entry, constructing a translation mapping relationship between the newly added translation data and the source entry; according to the translation mapping relation, cross-software to-be-processed source entry checking is carried out in the domestic operating system, and translation data updating or newly adding is carried out on the software with the detected to-be-processed source entry; and performing translation quality evaluation on the newly added translation data, and outputting an evaluation result of translation correctness.