119 results about "Adnoun" patented technology

Techniques for determining sentence based interactive topic-based summarization are provided. A text to be summarized is segmented. Discrete keyword, key-phrase, n-gram, sentence and other sentence constituent based summaries are generated based on statistical measures for each text segment. Interactive topic-based summaries are displayed with human sensible omitted text indicators such as alternate colors, fonts, sounds, tactile elements or other human sensible display characteristics useful in indicating omitted text. Individual and / or combinations of discrete keyword, key-phrase, n-gram, sentence, noun phrase and sentence constituent based summaries are dynamically displayed to provide an overview of topic and subtopic development within a text. A hierarchical and interactive display of texts based on the use of discrete sentence constituent based summaries which associates expansible and contractible displayed text provides contextualized access to an interactive topic-based text summary and to an original text.

This invention utilizes a concept called color, which implies a variation, and applies it to natural language attributes like verbs and nouns. The verb color is defined as a role or operation in which the field participates. The noun color is defined as a form of the field. The auxiliary verb color is defined as a path to the field from a known object reference. The noun color may be defined by the user or may be determined, based on the object state in which the field resides. Equations are made generic, by making the colors of the fields parametric. The equivalence of methods and equations was established, and a method might be invoked, as if it was a rule. Similarly, processes involving several methods, and other rules, may be described by rules. A special kind of classes called conceptual classes were invented, which can project a subset of the fields of a class, as well as group and reorder a particular field found in several classes. Several objects called collaboration objects, may interact with each other in several cycles, and in each cycle, several methods of the objects are invoked. Each method may view the collaboration objects in a predefined order called collaboration sequence. Temporary variables created during processing may be stored in a global or local table, and may be assigned user defined or state based nouns. By utilizing all the mechanisms defined above, rules may be specified and evaluated in a generic manner.

The invention discloses an SVM based micro-blog emotion classification method fusing various kinds of emotion resources. The method includes the following steps: constructing relevant dictionaries including an emotion dictionary, a negation dictionary, and a degree adverb dictionary; performing pretreatment on different corpora, performing word segmentation and part-of-speech tagging on the corpora, and performing sentence structure analysis; comparing the segmented words and positive and negative dictionaries to acquire initial word polarity, comparing words ahead of emotion words and the word degree grade dictionary and the negation dictionary to acquire modifier weight, and multiplying the initial word polarity by the modifier weight to acquire emotion scores of each micro-blog; extracting features such as nouns, verbs, adjectives, positive and negative emotion words, degree adverb weights, emotion scores, privatives and specific symbols from part-of-speech features, emotion features, sentence pattern features, and semantic features; and inputting the extracted features into an Libsvm to perform model training so as to acquire a training model. The method can achieve emotion 5-grade classification of micro-blogs, and can accurately and roundly acquire emotion tendency of netizens.

A useful object-oriented knowledge base system is provided, which comprises an 'object-oriented knowledge base', an inference mechanism, and an ideal dictionary, etc. Sentences used as a 'rule' and / or as a 'fact' in the 'object-oriented knowledge base' are described according to a simple English grammar. Hierarchical structure of nouns-system in an 'ideal thesaurus' is constructed, on the basis of special kind of 'object-oriented-lexical-definition of nouns' recorded in the ideal dictionary. Lexical meaning of a verb whose meaning is specific are derived from that of a verb whose meaning is general and universal, by using 'dichotomy' on the basis of C-language-like way of description of English sentences in the lexicon. The hierarchical structure of verbs-system in an 'ideal classification table' is constructed on the basis of them. The Inference mechanism processes not only mathematically well defined equations but, also simple English sentences, by making full use of the 'ideal thesaurus' and the 'ideal classification table', on the basis of specially contrived 'sentence based object-oriented categorical syllogism'.

The invention provides a question classification method and a question classification device for an automatic question-answering system. A class reference table including fine granularity classes of noun and interrogative pronoun is preset. The method comprises the steps: calling a preset interface for executing the class marking operation to divide a received question into a plurality of word segmentations, and carrying out the fine granularity class marking for the word segmentations according to the class reference table to obtain a corresponding fine granularity class mark sequence; matching the fine granularity class mark sequence with a preset first-level classification mode so as to primarily determine the type of the question; if the primarily determined type is non-unique, matching the fine granularity class mark sequence with a preset second-level classification mode so as to secondarily determine the type of the question; if the secondarily determined type is non-unique, matching the fine granularity class mark sequence with a preset third-level classification mode so as to determine the type of the question at a third time, and determining the type to be the type of the question if the type determined at the third time is unique. By adopting the method, the question classification efficiency and accuracy can be improved.

A method for automatically analyzing the text of a document to generate verification questions to be administered to a user as a quiz for the purpose of verifying whether the user has read the document. Syntactic analysis is applied to statements (e.g. sentences) in the text to automatically generate various types of verification questions, including fill-in-the-blank, true / false, and multiple-choice questions. Nouns and proper nouns in a statement may be used to generate fill-in-the-blank questions; numerical values may be used to generate fill-in-the-blank, true / false and multiple-choice questions; and verbs, adjectives and adverbs may be used to generate true / false questions. The questions may be generated dynamically for each user, or generated once, stored and used for multiple users.

The invention discloses a synonym expansion method and a synonym expansion device both used for text duplication detection, which include a text preprocessing unit used for deleting stop words in a suspected text and tagging the part-of-speech, wherein verbs, nouns, and adjectives are taken as the to-be-processed objects; through retrieving synonyms of single words, computing the Cartesian product and obtaining the initial expansion set of all word collocations in the suspected text; through comparing the initial expansion set and an actual corpus, filtering word collocations impossible in an actual language environment, simplifying the set, and obtaining the final expansion set; and during the duplication detection, according to different collocation results, giving the words different weights which are taken as the computation base for the duplication detection results. Through applying the method or the device disclosed by the embodiment of the invention, the problem of synonym replacement in text duplication can be efficiently overcome, the efficiency is higher, and the accuracy of the duplication detection is greatly improved.

Abbreviations are displayed for user-entered text commands, to facilitate discovery of keyboard shortcuts and to reinforce branding. Users enter commands by typing them into a text input field. Commands can be provided in a verb-noun structure, where the verb specifies what is to be done and the noun specifies the object or a parameter for the verb. Upon user entry of a command, or portion thereof, the entered portion is replaced by an abbreviation. The abbreviation can represent a single key, key combination, or multi-character string. The abbreviation can also include a logo or other graphic component, if desired. The abbreviation can replace the verb portion of the entered command, or it can be shown alongside or adjacent to the text input field, or it can be shown in an overlay or according to any other mechanism. A transition effect can be performed when introducing the abbreviation.

The present invention discloses a density-based text clustering algorithm research method. The method comprises the following steps: using the ICTCLAS word segmentation system to carry out word segmentation on a text in a text set, and extracting corresponding keywords from the word segmentation according to the three parts of speech of the noun, the verb, and the adjective, and the word frequency; using an improved HowNet word similarity algorithm to calculate keyword similarity of the obtained keywords; according to the keyword similarity in the text, calculating text similarity; and according to the obtained text similarity, using the density-based clustering algorithm to carry out clustering on the text, so that the performance of the existing text-related information retrieval technology can be significantly improved.

Techniques are disclosed for improved autonomous agents that can provide a recommendation in a non-intrusive, conversational manner. In an aspect, a method determines a first sentiment score for a first utterance and a second sentiment score for a second utterance, each sentiment score indicating an emotion indicated by the respective utterance. The method further identifies that a difference between the first sentiment score and the second sentiment score is greater than a threshold. The method further extracts a noun phrase from the second utterance. The method identifies a text fragment that includes an entity that corresponds to the noun phrase. The method identifies that the text fragment addresses a claim of the second utterance. The method forms a third utterance that includes the a recommendation related to the second utterance and adds the third utterance to the sequence of utterances after the second utterance.

The invention relates to the field of artificial intelligence, and provides a question and answer corpus generation method and device based on a text generation model, computer equipment and a storagemedium. The method comprises the steps of obtaining historical questions and a standard document, extracting keywords in the standard document and paraphrasing sentences corresponding to the keywords, performing word segmentation processing on the historical questions, identifying and discarding entity nouns in the historical questions to obtain syntactic feature words of the historical questions, combining the syntactic feature words with the keywords, and inputting the combined data into a pre-trained text generation model to obtain a target question corresponding to the keyword, wherein the text generation model by training based on a training sample marked with the keyword and syntax feature words are obtained, and according to the target question corresponding to the keyword and a paraphrasing statement corresponding to the keyword, a question-answer pair comprising the target question sentence and the paraphrasing sentence is constructed so as to improve the quality of the target question sentence and the question-answer pair.

A method of classifying and accessing writing examples for writing composition. A language domain is first selected and representatives texts from that domain are analyzed to build a classification system for the domain. The text is first analyzed to determine root nouns and root verbs. The texts are further analyzed to determine relationships between nouns and the root verbs used for each noun-to-noun relationship. At this point, writing examples are then extracted from the texts and stored in a database. These writing examples are then classified by the earlier defined noun-to-noun relationships and root verbs that go along with those noun-to-noun relationships. Access to the writing examples is accomplished via a three-level interface. The first level (noun interface) maps nouns and pre-determined relationships between those nouns. By selecting one of these relationships, a navigation link takes the user is a second level (verb interface) showing root verbs that relate to the particular noun-to-noun relationship selected. Here the user selects a particular root verb which causes a query of the writing examples database. The results of the query are sent to a third level interface (results interface) where the writing examples are displayed. The user may then select one or more writing examples to insert in a word processing program or document where the user may modify them for the writing job at hand.

The invention discloses a method of retrieving English text based on the matching degree. The method includes the following steps that firstly, in a server, retrieval information is stored in advance, each English document is related to a retrieval unit, wherein each retrieval unit includes ID, English document entry time and at least one search strip, each retrieval strip comprises at least one noun and notional verb in the abstract of the English document associated with the retrieval unit, and weights are preset for all the search strips; secondly, the retrieval of English is divided into noun and notional verb after inputting, the noun and the notional verb are expanded into retrieval statements; thirdly, retrieval the weight is obtained from the similarity evaluation of the retrieval statement, the retrieval weight is matched with the preset weight, results are sorted according to the matching degree and a list of retrieval results is obtained.

A phrase translation and language instruction system is provided comprising a plurality of printed sheets having disposed thereupon information relating to the translation of certain generic phrases. Additionally, there is provided therein a set of alternative verbs, nouns, adjectives and the like to enable the substitution of words relating to one's specific situation and the communication of such in a foreign language.

A system, method, and computer program product for generating a polar vocabulary are provided. The method includes extracting textual content from each review in a corpus of reviews. Each of the reviews includes an author's rating, e.g., of a specific product or service to which the textual content relates. A set of frequent nouns is identified from the textual content of the reviews. Adjectival terms are extracted from the textual content of the reviews. Each adjectival term is associated in the textual content with one of the frequent nouns. A polar vocabulary including at least some of the extracted adjectival terms is generated. A polarity measure is associated with each adjectival term in the vocabulary which is based on the ratings of those reviews from which the adjectival term was extracted.

The embodiment of the invention provides an information processing method and related equipment, which not only can determine the theme of a text, but also can obtain relatively long keywords and phrases corresponding to the text, and are richer in meaning, high in readability and more helpful to data analysis. The method comprises the steps of obtaining a target text; preprocessing the target text to obtain a target corpus set; inputting the target corpus set into a preset topic model to determine a topic corresponding to each word in the target corpus set; determining the theme of which theword frequency is greater than a second preset threshold value in the target corpus set as the theme of the target text; determining a target sub-tree according to a short statement method tree corresponding to the target text; combining nouns in the first sub-tree to obtain a keyword group corresponding to the target text; and determining the keyword group of which the word frequency is greater than a third preset threshold value in the keyword groups corresponding to the target text as the keyword group of the target text.

The invention relates to an associated word searching method and device based on semantics, and belongs to the technical field of computers. The semantic-based associated word searching method comprises the following steps: acquiring a text document from an internet database; identifying the text document by using a deep learning entity identification model to obtain an entity noun and an entity noun position; calculating a word vector of the entity noun according to the text of the sentence context of the entity noun determined by the entity noun position; performing word formation analysis on the entity nouns to determine entity types of the entity nouns; and according to the word vectors of the entity nouns and the entity types of the entity nouns, performing similar retrieval in a word vector library to search similar entity nouns. According to the semantic-based associated word searching method, the problem that one word is polysemy is solved, the method does not depend on an existing word bank, and unknown entity nouns can be processed.

The invention provides a text matching model training method and device, equipment and a storage medium, and the method comprises the steps: obtaining a plurality of first sample pairs, each first sample pair comprises two texts and a matching result of the two texts, carrying out the mask processing of entity nouns and / or verbs contained in the texts in each first sample pair, obtaining corresponding second sample pair, performing iteration pre-training on the initial text matching model based on the second sample pair to obtain a pre-trained text matching model, obtaining a plurality of third sample pairs corresponding to the preset business scene, wherein each third sample pair comprises two texts and a matching result of the two texts, based on the third sample pairs, performing iterative fine tuning training on the pre-trained text matching model to obtain a text matching model corresponding to the preset business scene. According to the invention, text matching can be carried out more accurately.

The invention relates to an artificial intelligence technology, and discloses a language model training method, which comprises the following steps of: respectively carrying out word-level mask, phrase-level mask, entity-level mask and part-of-speech-level mask processing on texts in a training data set to obtain a to-be-used pre-training data set; performing sentence vector representation processing on texts in the to-be-used pre-training data set to obtain a pre-training data set represented by sentence vectors; and inputting the pre-training data set represented by the sentence vector into the language model, carrying out model reasoning iteration training on the language model, and when a preset model training completion condition is satisfied, completing the training of the language model. The invention further relates to a block chain technology, and the training data set is stored in the block chain. The problems that in the prior art, a model obtained through an existing model training mode cannot learn information of a Chinese semantic level and information of a Chinese entity relation, and the sensitivity and accuracy of the model to nouns are low can be solved.

The invention discloses a method for generating an interactive scene in landscape conversion. The method comprises the step 1, establishing an initial scene state set to record initial scene information;step 2, establishing an advanced semantic operation set; step 3, performing natural language processing on the interactive text to generate an operation sequence; step 4, analyzing the operation sequence, and determining the operation type, the spatial relationship of the object, the attribute and the like; step 5, eliminating fuzzy semantics of the related indication pronouns and the collection nouns in the interactive text, and deriving a defective or implicit entity and spatial relationship in the interactive operation in combination with the initial scene and spatial knowledge; 6, reconstructing the initial scene according to the interactive operation; and step 7, generating a new scene by combining the 3D model library and the spatial knowledge base. According to the method, the problems of lack of user interaction and feedback and single scene in the current landscape conversion research are solved.

The invention belongs to the technical field of Internet. The invention relates to a method for discovering important noun labels through machine learning and context part-of-speech. The method comprises the following steps: S1, determined new words with certain data are listed through corpora, wherein the corpora need the most recent article, the articles of the last year are chosen as the corpus, the corpora are used for learning adjectives, connected words, verbs and the like in previous and later texts with the most possible nouns, and dozens of determined important nouns (such as Huawei,Xiaomi, ZTE, etc.) are sorted; and S2, through the operation of the previous step, a batch of collated sentence patterns such as 'XXX publishes... in the current year', 'no matter how XXX achieves...', 'for XXX' and the like can be obtained, and meanwhile, different probability situations can be obtained by calculating different sentence patterns. According to the method, the new words can be found through machine learning with extremely low manual intervention and arrangement cost.

The embodiment of the invention discloses a vulnerability information matching processing method and device. The method comprises the following steps: obtaining vulnerability related information in anetwork, carrying out the part-of-voice tagging and block extraction of vulnerability related information, and obtaining preprocessed vulnerability information; combining a plurality of blocks conforming to a preset grammatical structure in the preprocessed vulnerability information into a new noun block to obtain block vulnerability information; and matching verbs in the block vulnerability information according to preset sensitive verbs, and determining a target noun connected with the matched target verbs as vulnerability information. Part-of-voice tagging and block extraction are carried out on vulnerability related believes in a network; according to the vulnerability information matching method and device, the blocks conforming to the combined grammatical structure of adjectives andnouns are combined, so that each block can be quickly matched when vulnerability information matching is performed, the vulnerability information matching efficiency is greatly improved, the labor cost is reduced, Meanwhile, vulnerability processing in time and loss reduction are facilitated.

The invention discloses an English grammar training method and an English grammar training system. The English grammar training method comprises the following steps: acquiring text information and audio and video files corresponding to the text information from a server; identifying the part-of-voice of each word in the text information, the part-of-voice being a word of a noun and being marked asa first type of words; wherein the part-of-voice is a word in a word segmentation form of verbs and is marked as a second type of words; wherein the part-of-voice is a word of a verb, does not include an emotional verb and a second type of words, and is marked as a third type of words; displaying one or more statements, wherein the first type of words, the second type of words and the third typeof words are all hidden; playing an audio and video file corresponding to the statement; responding to external input of a user; judging whether the external input of the user is matched with the first type of words, the second type of words or the third type of words or not, if yes, displaying the words, and if not, waiting for the user to input again and judging again.

A data processing method and apparatus are disclosed. The method includes obtaining all the nouns in text information to take all the nouns as entities; determining an entity type corresponding to each of the nouns; analyzing the text information and determining a grammatical relationship type between the entity types corresponding to the nouns having a grammatical relationship to regard the grammatical relationship type as the relationship between the entities. When the method is adopted, the relationship between the entities in the text information can be obtained, so as to facilitate subsequent processing by using the obtained relationship between the entities.

The invention relates to a key phrase generation method based on a training model. The method comprises the following steps: S1, obtaining to-be-processed text data; S2, performing word segmentation and part-of-speech tagging on the acquired text data; S3, establishing a disabled lexicon, and removing words existing in the disabled lexicon; filtering out words which are not verbs and nouns; S4, performing N-gram combination to obtain a candidate word combination; S5, performing text vector conversion on the text data and the candidate word combination based on a pre-training model of Bert; S6, performing cosine similarity calculation on the vector representation of the document level and the vector representation of the candidate word, and performing semantic similarity sorting; and S7, according to a set value, selecting the words or phrases with the semantic similarity ranks in the top in the step S6 to form keywords. According to the method, the open-source pre-training model Bert is used for carrying out text vectorization expression, information of the semantic level of the text is completely obtained, keyword extraction is facilitated, phrase-level keywords are obtained according to N-gram combination, and the meaning is more complete compared with single words.

The invention provides a barrage text clustering method based on feature extension and T-oBTM. The method comprises three steps of a network new word processing stage, a theme modeling stage and a text clustering stage. The invention provides an oBTM streaming short text clustering method (T-oBTM) for carrying out threshold constraint on word pairs according to bullet screen characteristics, the algorithm execution time is shortened, network new words are recognized and processed, the purpose of expanding text characteristics is achieved, and then the algorithm precision is improved. Accordingto the method, the network new words are recognized and processed, the word segmentation lexicon is enriched, and the word segmentation precision is improved; when the network new words are processed, the recognized entity nouns and sentiments, viewpoints and opinion words are processed differently, short text features are expanded, and clustering precision is improved.

A text topic mining method based on word semantic weight of Internet service comprises the following steps: 1, performing part-of-speech tagging on words in a Mashup service description document by using a natural language toolkit in Python; 2, counting word frequency information, and calculating TF-IDF information; 3, extracting Mashup service label information wherein the semantic weight of each word in the Mashup service description document is recalculated on the basis of the noun set Nset and the TF-IDF value; and 4, solving Mashup theme features through an NMF model. On the basis of TF-IDF, in combination with service label information and context word information, weights of words are recalculated, and weight values of key words are increased, so that Mashup service modeling and service document theme confirmation are effectively performed.

The invention discloses a hybrid information extraction method and system oriented to an open domain. According to the method, firstly, through context clause decomposition and NLP preprocessing, a composite sentence is simplified, and language attributes of the sentence are obtained; then identifying explicit phrases in the sentences, and identifying implicit phrases by using the two defined extension rules; and finally, based on the defined six language scene rules or the combination thereof, extracting the relationship between the identified entities, and generating a relationship triple. Aiming at the defects of non-atomic extraction, implicit phrases in a text are expanded by utilizing dependency analysis and heuristic rules, long noun phrases and relation phrases can be further decomposed into meaningful and more compact phrases, and the atomic requirement of information extraction can be better met; for cross-clause extraction, co-reference resolution and context clause decomposition are adopted for compositing sentences, the relation on the simplified clauses is extracted in combination with dependency analysis, and the extraction capacity on a composite sentence pattern is improved.