专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

AU2021104218A4 A SYSTEM FOR IDENTIFICATION OF PERSONALITY TRAITS AND A METHOD THEREOF 未知
公开(公告)号：AU2021104218A4
公开(公告)日：2021-09-09
申请号：AU2021104218
申请日：2021-07-16
申请人： BAGGA JASPAL , CHOPRA JHARNA , DEWANGAN AMIT KUMAR , KUMAR NAGWANI NARESH , KUMAR KALLEPALLI ROHIT , SHARMA SURAJ , SRIVASTAVA SUMIT , VERMA AMRITA
发明人： SHARMA SURAJ , SRIVASTAVA SUMIT , BAGGA JASPAL , KUMAR NAGWANI NARESH , DEWANGAN AMIT KUMAR , KUMAR KALLEPALLI ROHIT , CHOPRA JHARNA , VERMA AMRITA
IPC分类号： G06F40/00 , G06K9/00 , G06N3/02
摘要： A system for identification of Personality Traits in Text, comprises of: an input module 102 for receiving a plurality of text data, wherein the plurality of text data is segregated into a plurality of datasets based on label, wherein the plurality of dataset is divided into a training and a validation dataset, an extraction module 104 for extracting the plurality of text data from the plurality of partitioned dataset, wherein the plurality of text data is pre-processed using a time efficient sentence tokenization algorithm, an encoding module 106 for encoding each word of the text using an encoding technique, wherein a document formed of the encoded text undergoes padding and shortening of the data and a training module 108 for identifying the personality traits from the encoded text using a neural network, wherein an optimization module optimizes a learning rate for obtaining a precisely identified personality trait. INPUT MODULE EXTRACTION MODULE ENCODING MODULE TRAINING MODULE receiving a plurality of text data an input module 102, wherein the plurality of text data is segregated into a plurality of datasets based on label, wherein the plurality of dlataset is divided into a training and a validation dataset 202 extracting the plurality of text data from the plurality of partitioned dataset using an extraction module 104 connected to the input module 102, wherein the plurality of text data is pre-processed using a time efficient sentence tokenization algorithm 204 encoding each word of the text using an encoding technique using an encoding module 106 connected to the extraction module 104, wherein a document formed of the encoded text undergoes padding and shortening of the data 206 identifying the personality traits from the encoded text using a neural network of a training module 108 connected to the encoding module 106, wherein an optimization module optimizes a learning rate ofthe training module 108 for obtaining a precisely identified personality trait 208

IPRDB

热门服务

关于我们

友情链接

联系方式