US 11,775,751 B2Granted

Method and system for extracting valuable words and forming valuable word net

AssigneeAwoo Intelligence, Inc.

InventorsKuo-Ming Lin, Szu-Wu Lin

Life of this patent

Estimating the term…

Loading patent…

Abstract

Method and system for extracting valuable words and forming a valuable word net, wherein a server is employed to collect contents of articles from internet sources, EDM texts (email direct marketing), product descriptions and other texts, and extract valuable words by machine learning. Each valuable word performs connection weight based on the number of times the text has been read, the number of times the text has been clicked, the number of times the text has been cited, the correlation of external sites, the conversion of expert knowledge, the probability space, the Shannon entropy, the spatial distribution and other values and their conversion by machine learning; then the linked valuable words are integrated to form a valuable word net. When the valuable words are needed to be used, the valuable words and the valuable word net can be retrieved from the database for subsequent applications.

Claims12

Dependent claims

A method for extracting valuable words and forming a valuable word net, comprising following steps:

collecting texts, wherein a server is connected to third-party search system via internet, and independently collects one or more text information and text-related information related to the text information through the third-party search system;
training and learning the texts, wherein the server continuously collects and transmits the text information and the text-related information to a word extraction module of the server, and the word extraction module uses the text information and the text-related information as the first training data to perform a first machine learning;
extracting valuable words, wherein, after the word extraction module completes the first machine learning, the word extraction module extracts one or more valuable words from the text information based on a first machine learning result, and stores the valuable word in the server;
learning/training the valuable words, wherein a word net linking module of the server stores one or more of the valuable words and the text-related information as a second training data for a second machine learning;
linking the valuable words, wherein, after the word net linking module completes the second machine learning, the valuable word is linked to another valuable word or a plurality of the valuable words with a word weight based on a second machine learning result;
and
forming a valuable word net, wherein the word net linking module integrates one or more groups of the valuable words linked based on the word weight into the valuable word net, and stores the valuable word net in the server.

4 more elements

A system for extracting valuable words and forming a valuable word net, comprising:

a server comprising a data processing module which is respectively connected to a data storage module, a data collection module, a word extraction module, and a word net linking module, wherein the data processing module is employed to operate the server;
wherein the data storage module comprises a learning database, a valuable word database, and a word net database;
wherein the data collection module collects and stores a text information and a text-related information in the learning database;
wherein the word extraction module makes use of the text information and the text-related information stored in the learning database as a first training data to perform a first machine learning, a valuable word is extracted from the text information based on a first machine learning result, and the valuable word is stored in the valuable word database;
wherein the word net linking module uses the valuable word and the text-related information as a second training information and performs a second machine learning, the valuable word is linked to another valuable word or a plurality of the valuable words according to a word weight based on a second machine learning result, and then the word net linking module generates one or more groups of the valuable words linked based on the word weight into a valuable word net, and the valuable word net is stored to the word net database;
and
a third-party search system providing the text information and the text-related information to the server.

4 more elements

Dependencies are inferred from the claim text.