Automated Text Analysis Based on Skip-Gram Model for Food Evaluation in Predicting Consumer Acceptance

Algorithm 1

Algorithm for taste and smell analysis.

Algorithm. Find_Taste_and_Smell(Keyword, Corpus, n)

Input: String Keyword and List Corpus of n integers

// keyword is “Mat (Taste)” or “Hyang (Smell)”

Output: List Result of most similar words in Corpus

// Corpus is have to be jjampong ramen A corpus or jjampong ramen corpus B

Method:

  begin

    Result ← an empty List

    vector ← wor2vec(keyword) // represent input string A as a vector by skip-gram model

    for ← 0 to

      word_vector ← word2vec(Corpus[])

      similarity ← cosine_similarity(vector, word_vector) //compute similarity

      Result.insertElem(word, similarity)

    end

    sorted(Result) // sort descending by similarity and store only top 20 words

    filtering(Result) // remove noise words

    return Result

  end