Abstract: Method, apparatus, and non-transitory storage medium for training two or more cross-task neural network models based on two or more neural network tasks, including mapping first pseudo labels based on a first model associated with a first task among the two or more neural network tasks and second pseudo labels based on a second model associated with a second task among the two or more neural network tasks to a same space, and computing a matching score indicating a cross-task matching between the first pseudo labels and the second pseudo labels based on the mapping. The method may further include selecting one or more cross-task pseudo labels based on the matching score and accuracies associated with the first model and the second model, and training the two or more cross-task neural network models based on the one or more cross-task pseudo labels.

Abstract: There is included a method and apparatus comprising computer code for a joint training method using neural networks with noise-robust losses comprising encoding input tokens from a noisy dataset into input vectors using an input encoder; predicting a label based on the input vectors using a classifier model; calculating a beta value based on the input vectors and the label using a label quality predictor model, wherein the beta value is instance-specific for each training instance; and j oint training more than one model using a first modified loss function based on the beta value and an entropy value.

Abstract: There is included a method and apparatus comprising computer code configured to cause a processor or processors to perform generating one or more aligned inventories, wherein the one or more aligned inventories are generated using one or more word sense inventories, obtaining a word in a context sentence, determining one or more semantic equivalence scores indicating semantic similarity between the word in the context sentence and each of one or more associated glosses in the one or more aligned inventories using a semantic equivalence recognizer model, and predicting a correct sense of the word in the context sentence based on the determined one or more semantic equivalence scores.

Abstract: A method and system for selecting an encoding format used for reading a target document are provided. The method comprises reading a reference document with at least one reference encoding format and determining all or some disorder code patterns obtained when reading the reference document with the reference encoding format; reading the target document with one encoding format each time; for each encoding format, comparing data generated when reading the target document with this encoding format and the determined disorder code patterns to determine disorder codes generated when reading the target document with this encoding format; counting disorder codes generated when reading the target document with each encoding format, and making a comparison to determine the encoding format used for reading the target document.

Abstract: A method and system for key knowledge point recommendation are provided, the method comprising calculating knowledge point relationship strengths of knowledge points in a set of knowledge points; calculating weights for knowledge points according to the knowledge point relationship strengths of knowledge points in the set of knowledge points, and storing the knowledge points and weights correspondingly; determining key knowledge points according to the weights of the knowledge points and recommending the key knowledge points to a user.

Abstract: The present invention provides a semantic information acquisition method and system, and corresponding keyword expansion and search methods and systems, comprising: searching for, then classifying an article; then, performing word segmentation according to the classified article to obtain the words in said category, and setting said category and words to serve as the semantic information of the keyword; also, a method and system using the semantic information acquisition method to expand a keyword, and a method and system using keyword expansion to perform a search. The described semantic information acquisition method effectively avoids the technical problems in the prior art of only being able to obtain semantic information of English vocabulary; and it also being impossible to classify semantic information based on category information. The invention is particularly suitable for searching using a keyword, searching a large number of texts, and organizing large amounts of related data and information.

Abstract: A method and system for obtaining a knowledge point implicit relationship are provided; first, establishing a knowledge point explicit relationship map according to knowledge point explicit relationship strengths; second, computing according to said knowledge point explicit relationship map a simple path set of two knowledge points; then, computing the implicit relationship strength values corresponding to each simple path in said simple path set; further, comparing the relationship strength values of the simple paths and setting as the significant implicit relationship strength value the simple path relationship strength having the largest value also greater than a preset threshold value.

Abstract: A method and system for selecting an encoding format used for reading a target document are provided. The method comprises reading a reference document with at least one reference encoding format and determining all or some disorder code patterns obtained when reading the reference document with the reference encoding format; reading the target document with one encoding format each time; for each encoding format, comparing data generated when reading the target document with this encoding format and the determined disorder code patterns to determine disorder codes generated when reading the target document with this encoding format; counting disorder codes generated when reading the target document with each encoding format, and making a comparison to determine the encoding format used for reading the target document.

Abstract: A keyword expansion method and system are provided. The method comprises searching with a predetermined initial keyword to obtain current keywords used as a basis of a next search, performing loop search through keyword iteration. If a keyword error between keywords obtained in the current search and those keywords obtained in a previous search is less than a predetermined threshold, using the keywords obtained in the current search as expanded keywords of the initial keyword. With this method, the problem of manually establishing a thesaurus in the prior art may be solved. A method and system of automatically annotating a classified corpus is also provided. The method comprises: determining one or more initial core keywords for each class; obtaining expanded keywords for each class through expanding the initial core keywords; searching with the expanded keywords corresponding to a class to select a classified corpus and annotating the classified corpus.

Abstract: A method and system for key knowledge point recommendation are provided, the method comprising calculating knowledge point relationship strengths of knowledge points in a set of knowledge points; calculating weights for knowledge points according to the knowledge point relationship strengths of knowledge points in the set of knowledge points, and storing the knowledge points and weights correspondingly; determining key knowledge points according to the weights of the knowledge points and recommending the key knowledge points to a user.

Abstract: In the method and system for knowledge extraction of this invention, knowledge extraction is realized through acquiring an initial sentence group including one or more sentences, and then comparing the length of the initial sentence group with an expected length to determine the initial sentence group to be expanded according to the comparison result. Since the sentence groups are formed by consecutive sentences, it may be guaranteed that the sentence groups themselves have good coherence in logic, so that the final sentence groups obtained through expanding the initial sentence groups have good coherence in logic correspondingly. Thus, this invention may override the drawback of lacking logical coherence in extracted knowledge information in the prior art.

Abstract: The present invention provides a semantic information acquisition method and system, and corresponding keyword expansion and search methods and systems, comprising: searching for, then classifying an article; then, performing word segmentation according to the classified article to obtain the words in said category, and setting said category and words to serve as the semantic information of the keyword; also, a method and system using the semantic information acquisition method to expand a keyword, and a method and system using keyword expansion to perform a search. The described semantic information acquisition method effectively avoids the technical problems in the prior art of only being able to obtain semantic information of English vocabulary; and it also being impossible to classify semantic information based on category information. The invention is particularly suitable for searching using a keyword, searching a large number of texts, and organizing large amounts of related data and information.

Abstract: A method and system for obtaining a knowledge point implicit relationship are provided; first, establishing a knowledge point explicit relationship map according to knowledge point explicit relationship strengths; second, computing according to said knowledge point explicit relationship map a simple path set of two knowledge points; then, computing the implicit relationship strength values corresponding to each simple path in said simple path set; further, comparing the relationship strength values of the simple paths and setting as the significant implicit relationship strength value the simple path relationship strength having the largest value also greater than a preset threshold value.

Abstract: A translation and teaching system for sending text messages over a telecommunication pathway sent by users in a first language received by a server, a comparison database of a second language for matching, converting the text into a second language upon making the match, a translator for converting the first text transmission into the second language if no match is made into a second transmission, a processor in the server for analyzing the conversion of the transmission for accuracy, and a data path to send the transmission to the user. The system may have a pre-processor that recognizes linguistic irregularities and modifies them to conform to the second language's pre-stored standard grammar and a human linguist platform for post verification of the translation. The system also generates learning lessons for user in language, a learning dictionary that groves with user input, and the addition of third party content for lessons.