Extragerea unui sentiment uman dintr-un text folosind o rețea neuronală recurentă și biblioteca Keras

Revista Română de Informatică și Automatică

卷:30

Paul TEODORESCU¹ 

[1]Institutul Naţional de Cercetare-Dezvoltare în Informatică – ICI București

关键词: library;    vector;    tensor;    matrix;    lstm cells;    labels;    variable;    back propagation;    forward propagation;   

DOI  :  10.33436/v30i3y202009

来源: DOAJ

【 摘 要 】

: In this paper, it is proposed to understand how the computer is able to extract a simple humanfeeling of "liked" or "disliked" from a text. Basically the computer will learn to correctly place a moviereview in one of the two categories of positive or negative. We’ll see how, starting with input values andoutput values called labels, the computer begins to learn and correctly recognize the output value (in this casethe 0 or 1 digit, zero representing a negative feeling and the one a positive feeling) through a model built onthe technique called supervised learning. So the proposed objective is to guess the human feeling (translatedby the number 0 or number 1) which is in fact the output value of the model, at a new value of the input, oncethis model has been known. In this exercise we will use Keras API built on TensorFlow, a set of moviereviews taken from IMDB and a recurring neural network RNN with LSTM (Long-Short Term Memory)cells to preserve the memory of the words that were previously encountered. Keras comes with a set of50,000 movie reviews that were already pre-processed (this will be explained below). By feeding the neuralnetwork with these tens of thousands of texts (25,000 texts for training followed by another 25,000 texts fortest), the model built by Keras (using relationships of the words), manages to guess with a good accuracy, thepositive or negative human feeling, in other words the polarity of the text. The applications for sentimentanalysis are endless starting from social media monitoring and VOC, tweets and facebook posts analyzes, tothe business analysis by text analysis.

【 授权许可】

Unknown