Abstract | ||
---|---|---|
Companies, among other sectors, require that the opinions generated on the web be extracted automatically, obtaining their polarity on products or services, to achieve their objectives. Since the opinions are subjective and unstructured, there are still many problems within this field that must be solved. To mention a few, the problem of ambiguity and the support of languages, directly affect in the time to make the right classification of opinions, because most of the tools used in the processing of texts, they only work well with data in English. With the aim of contributing to the solution of both problems and evaluating the real behavior of sentiment analysis for the Spanish language, a system is proposed that allows determining the positive or negative polarity, trying to detect the irony as a problem of ambiguity. For the classification, a supervised learning method was implemented, with the Naive Bayes algorithm. The evaluation of the results of the classification shows that the problem of detecting ironies in Spanish, using the classical techniques of opinion mining, is not completely resolved. However, we believe that these results can be improved by applying some strategies. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1007/978-3-030-20055-8_49 | 14TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING MODELS IN INDUSTRIAL AND ENVIRONMENTAL APPLICATIONS (SOCO 2019) |
Keywords | Field | DocType |
Opinion mining,Sentiment analysis,Social networks,Natural language processing,Text mining,Twitter,Classification algorithms | Irony,Social network,Naive Bayes classifier,Sentiment analysis,Computer science,Supervised learning,Artificial intelligence,Natural language processing,Statistical classification,Ambiguity,Machine learning | Conference |
Volume | ISSN | Citations |
950 | 2194-5357 | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Daniela E. Sanjinés | 1 | 0 | 0.34 |
Vivian F. López Batista | 2 | 63 | 12.89 |
Ana B. Gil | 3 | 49 | 14.81 |
María N. Moreno García | 4 | 148 | 26.06 |