Method for classifying easy-to-confuse categories based on two-stage model

• Main Authors: WANG CHANGJIAN, ZHANG YUANLIANG, FANG CUNDI, LI JIAYING, CHEN LIQIAN, JIANG ZHIJIE, DONG WEI, LI SHANSHAN, MA YINGWEI, LI XINGPEI
• Format: Patent
• Language: Chinese; English
• Published: 24.11.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a method for classifying easy-to-confusion categories based on a two-stage model. According to the technical scheme, a two-stage classification system composed of a preprocessing module, a first-stage classification module, an easy-to-confusion category recognition module, a boundary sample screening module and a second-stage classification module is constructed. A training set and a test set are constructed, and a preprocessing module maps texts or source codes in the training set and the test set into a word-level coding sequence set; a first-stage classification module classifies the word-level coding sequence set according to a category table to obtain a confusion matrix; the confusion type identification module obtains a confusion type two-tuple sequence by using the confusion matrix; a two-stage classification system is trained, and the trained two-stage classification system performs preprocessing, first-stage classification, easy-confusion category identification and boundary s