Quantcast
Viewing all articles
Browse latest Browse all 63164

Feature Similarity Measures

Hello,

I have a question to ask you guys. Its regarding similarity measures that can be used to find similar features.

The features are text based. e.g.

sample 1-> word1 word2 word3 word4.... sample 2-> word2 word1 word4 word5.... sample 3-> word2 word1 word4 word5.... 

As we can see, the similarity should be in terms of the words and their order is very important. From the example, sample-2 and sample-3 are the most similar because of the order of the words. Any suggestions are much appreciated.

Is Cosine similarity good enough?

submitted by warriortux
[link][4 comments]

Viewing all articles
Browse latest Browse all 63164

Trending Articles