The invention discloses an algorithm of extracting text digest according to keywords. The algorithm comprises the following steps of S1. splitting a text into different paragraphs, and storing the paragraphs in a set; S2. for different paragraphs, determining initial positions of keywords in the paragraphs in sequence; S3. splitting the paragraphs according to keywords so as to obtain a paragraphset not including keywords; S4. extracting digest of a paragraph according to the keyword and lengths of to-be-extracted texts before and after the keyword; S5. putting the extracted digest into the set; and S6. determining whether the digest is extracted in each paragraph, if not, returning to S2, if yes, ending the algorithm. Through adoption of the algorithm, when a lot of texts are unnecessaryto read carefully or may cost a lot of time to read, people can extract a part of digest information according to keywords so as to read generally, or people can read carefully if the text is worth of reading. The algorithm of extracting text digest according to keywords is particularly applicable to extracting interested content in a lot of texts.

Algorithm of extracting text digest according to keywords

The invention discloses a book classification method. The method comprises the following steps of an obtaining step, wherein a classification database and book information of books to be classified are obtained, the book information comprises book content information, and the classification database comprises classification entries and key word groups corresponding to the classification entries; aword segmentation step, wherein word segmentation treatment is conducted on the book content information of the books to be classified to obtain word segmentation word groups; a classification determining step, wherein multiple keywords in the word segmentation word groups are obtained to form characteristic word groups, and characteristic words are compared with the key word groups to determinethe classification entries corresponding to the books to be classified. The invention further provides a classification database building method. According to the book classification method, by collecting the characteristic word groups in the books to be classified and comparing the characteristic word groups with the content in the classification database to determine the types of the books to beclassified, the manual classification defects of complicated documentation and consuming time are overcome, and the classification efficiency and accuracy are improved.

Classification database building method and book classification method

The invention provides an ancient book name identification method and apparatus. The method comprises the steps of obtaining an ancient text, and based on a word segmentation algorithm of an ancient book statistics model, performing word segmentation processing on the ancient text to obtain multiple segmented words; based on a pre-stored book name knowledge base, classifying the segmented words, wherein classification types include book names, chapter names, dance and music names, book name abbreviations, chapter name abbreviations and person name-containing book names; and for each segmentedword, when the segmented word is judged to only belong to one type in the classification types, storing the segmented word as a target book name of the ancient text. According to the method, the problems of time and effort consumption of conventional manual book name tagging and relatively poor effect of conventional machine tagging can be relieved; and the ancient book name identification precision and quality are improved, so that actual application demands in ancient book sorting and publishing work are met.

Ancient book name identification method and apparatus

The invention provides an ancient book proper name recognition method and device. According to the method, an ancient book text is acquired, word segmentation processing is performed on the ancient book text based on an ancient book statistical model word segmentation algorithm, and multiple proper names are obtained; each proper name is classified based on a pre-saved proper name knowledge base;for each proper name, whether the proper name is a book name in the pre-saved proper name knowledge base is judged; and if not, for each proper name, when it is judged that the proper name only belongs to one of classification categories, the proper name is saved as a target proper name of the ancient book text. Through the method, the problems that traditional manual proper name annotation consumes time and labor and traditional machine annotation has a poor effect can be relieved, the precision and quality of ancient book proper name recognition are improved, and therefore the actual application demand in ancient book sorting and publishing work is met.

Ancient book proper name recognition method and device

The invention discloses an automatic test method of an open platform page of a blogger cloud based on word segmentation retrieval, belonging to the technical field of the open platform of the Makeshift Cloud. Each page is designed as a Class, which contains the elements to be tested in the page, such as buttons, input boxes and titles. The page elements can be obtained by calling the page class inthe Selenium test page, and the page elements can be obtained by calling the page class. The page elements can be extracted according to the natural language input by the input template, and the sentence-breaking operation and judgment can be performed. The prior knowledge base established by the invention according to the machine learning converts the clauses judged to be correct in the form ofbroken sentences into machine language for the machine to self-identify and realize, and has the advantages of simplifying the work content of the software test engineer and improving the work efficiency of the software test engineer.

An automated test method for the page of the open platform of the blogger cloud based on word segmentation retrieval

The utility model relates to a saaS ancient book knowledge service cloud platform, the text data storehouse respectively with the network resource database, independently the collection of data storehouse is connected, the automatic word segmentation module of text data storehouse through being connected with the expert knowledge library respectively, the automatic classification module, the automatic indexing module is connected with digital ancient book archives, web retrieval module is connected to digit ancient book archives, digit ancient book archives includes three level, be the original data document databse respectively, concern the index document databse, the document databse is learnt to the logic, digit ancient book archives passes through the cluster module with information storage in original data document databse, digit ancient book archives loops through the feature recognition module, feature extraction module, dependence calculation module, the cluster module is with information storage in concerning the index document databse, the logic is learnt the document databse and is connected with logging module through the intelligent analysis module, it a make full use of saaS management mode's proper noun automatic indexing, the full -text search, intelligent extraction's saaS ancient book knowledge service cloud platform.

SaaS ancient book knowledge service cloud platform

The invention provides an ancient document management system and method based on a knowledge discovery technology. The system comprises an ancient document processing database, an ancient document import module, a database content management unit, a document information classification management unit and a retrieval part, and the retrieval part comprises a full-text retrieval unit and a knowledge discovery function retrieval unit. According to the ancient document management system and method based on the knowledge discovery technology, by means of an automatic sound indicating and intelligent extraction technology, the difficult problem of a word segmentation technology of a dictionary is solved, three modes of a reverse maximum matching method, a word segmentation method based on frequency statistics of words and a word segmentation method based on knowledge understanding are comprehensively adopted, word segmentation processing of estimated text is achieved, the speed is high, the error rate is low, meaningful and concise knowledge is extracted from ancient document data, data is converted to be of a proper format, a mode is found from the data, found results are evaluated and explained, the knowledge discovery is profound and accurate, and practical reference significance can be provided for a user.

Ancient document management system and method based on knowledge discovery technology

The invention relates to an associated word extraction method based on proper nouns The method comprises the steps that 1, a text set of a set field is acquired; 2, the text set is preprocessed to obtain feature texts; 3, an associated word of every two proper nouns is extracted from the feature texts; 4, according to relevant information of each of the associated words, regression analysis or belief network technology is used to find the interdependent relation between each associated word and the feature texts; 5, according to the interdependent relation, a part or all of the associated words are selected Through the associated word extraction method based on the proper nouns, a maximum amount of information can be acquired with least time

Associated word extraction method based on proper nouns

The invention provides an establishment method for an intelligent digital literature database of ancient books. The establishment method comprises the following steps: S1, using an automatic word segmentation technology to establish a digital archive of the ancient books which takes words as units; S2, using a computational intelligence technology, automatically classifying proper nouns through a clustering method, wherein the computational intelligence technology comprises a evolutionary algorithm, a neural network and fuzzy logic; S3, successively establishing intelligent databases of an original data literature database, a relational index literature database and a logic learning literature database which are in three levels, wherein in the clustering method, when classification criteria cannot be determined in advance, the fuzzy logic and swarm intelligence methods are used for groping a reasonable category template. Basic knowledge databases comprising databases related to the proper nouns of names, words, numbers and the like are established, knowledge is extracted from large amounts of data samples, data mining methods of the fuzzy logic and decision tree classification and the like are used to find out useful information of useful names, place names, events and like to achieve omnidirectional three-dimensional relationship retrieval.

Establishment method for intelligent digital literature database of ancient books

The invention belongs to the field of clustering methods and particularly relates to a method for clustering proper nouns of ancient books based on an evolutionary algorithm. The method for clustering the proper nouns of the ancient books based on the evolutionary algorithm comprises the following steps of step 1.feature extraction; step 2.classifcation; step 3.clustering; step 4.correlation analysis; and step 5.variance analysis, wherein the classification is as follows: some proper nouns are classified according to some attribute features of processed data, the similarity between various kinds of data is small, but the similarity degree between the data of one kind is large; after classification is completed, the data of each kind is identified by a unique mark, and the common characteristics of the data of the kind are also extracted for feature description of the kind. According to the method for clustering the proper nouns of the ancient books based on the evolutionary algorithm, which is provided by the invention, the full-text search and the knowledge discovery can be realized.

Yang Xiaoqiang

Papers

SaaS ancient book knowledge service cloud platform

Ancient document management system and method based on knowledge discovery technology

Associated word extraction method based on proper nouns

Establishment method for intelligent digital literature database of ancient books

Method for clustering proper nouns of ancient books based on evolutionary algorithm