RadostinaIvanova / Personality-test

Personality test which classifies in four personality types. For the classification is used the natural language processing classification algorithm - Multinomial Naive-Bayes.

Personality-test

Personality test which classifies in four personality types. For the classification is used the natural language processing classification algorithm - Multinomial Naive-Bayes. The Markov model is used as language model to expand a sentence with its context words for more precision and greater results.

What does it do?

Server asks questions, a person answers them and by the the given answers a trained Naive Bayes classificator classificates the person. To be specific The markov model gives for every answer context, concatenates it to the answer to expand the text for classification.

Multinomial Naive Bayes

The multinomial NB model, a probabilistic learning method. The probability of a document d being in class c is computed as

result = argmax(c∈ℂ)Pr[c|d] = arg max(c∈ℂ) log Pr[c] + ∑(from k = 1 to n)log Pr[t|dk]

N-gram Modeling With Markov Chains

Language model (in the narrowest sense) is a probability distribution over a sequence of words (such as a sentence) The Markov Model is implemented With Jelinek-Mercer interpolated smoothing:

Pr̂ int[ xk | x1x2... k - 1 ] = λ#(x1x...xk)/#(x1x2…xk−1∙) + (1 − λ)Pr̂ int[xk | x2x3...xk−1]

Install

1. Make sure you have GoLang installed on your local machine https://golang.org/doc/install

2. After you successfully install go, run the following command

   go get github.com/RadostinaIvanova/Personality-test

Running

In two different terminals in first run the server and in the second client. It can have multiple clients connected and using it at the same time.

go run server.go

go run client.go ui.go