(Recently, I was writing a paper for sentiment analysis, and used SVM as a classification algorithm for sentiment classification. I chose the famous SVM open source toolkit libSVM, which works well. Because the input corpus format of LibSVM has certain requirements. Therefore, sometimes how to convert everyone's practice corpus into the input corpus format of LibSVM is a laborious work. In the process of doing this, I also encountered a lot of troublesome office work. For example, at the beginning, I put samples of the same type in one, and the result was that libSVM could not be accurately classified. . . It took me a long time to figure out what the cause was, but it was finally cleared up. Later I will write about some experiences of using libSVM. I have sorted out the program for text classification based on SVM, and now I will share it with you, because the program uses ICTCLAS, a word segmentation tool of the Chinese Academy of Sciences, and LibSVM, please download it from their official website. As long as you follow the instructions of the first step of the readme text classification in my text classification program.txt, you can complete the text classification step by step. Because I am in a hurry to write the thesis, I will write a batch management file in the next stage to make it easy to use. This document includes the following attachments:
Text Classifier (using libSVM)\dict.txt
Text classifier (using libSVM)\featureselection.exe
Text Classifier (using libSVM)\getFeature.exe
Text Classifier (using libSVM)\getRandFile.exe
Text classifier (using libSVM)\getSVMTtrain.exe
Text Classifier (using libSVM)\mergeFile.bat
Text classification program (using libSVM)\readmeThe first process of text classification.txt
Text Classifier (using libSVM)\seg.exe
.....)
发表于 2022-9-24 16:27:01
粤ICP备14042591号-1
粤ICP14042591号