Name: Peng Xu Email pxu105@syr.edu
step 1: put the foods.txt(sorce text) and stopWordsList.txt in the right directory D:\Problem3...
step 2: run the code D:\Problem3\src\main\com\wordCount.java, the console will print the top 500 frequent words and the document most500FreWords.txt and reviewVextor.txt will be created at D:\Problem3..., the reviewVextor records all the reviews' vector data in readable format for weka, over 1Gb, you may not need to open it.
step 3: import weka.jar to the project, run the code D:\Problem3\src\main\com\KmeansByWeka.java, wait for around the 20 minutes, the console will print the final 10 clusters by K-means method by weka, the clusterResult.txt will be created then.
step 4: copy the data in clusterResult.txt to an excel, the you can get top 5 words of each cluster by selecting
The result text has already been created under directory D:\Problem3\result
Top500 frequent words in : Problem3\result\most500FreWords.txt
10 clusters by weka k-means: Problem3\result\clusterResult.txt
Top5 5 words of each clusters: Problem3\result\top5words.txt