Sentiment Detection

Sentiment Detection (also sentiment analysis , English for “mood recognition”) is a sub-area of text mining and describes the automatic evaluation of texts with the aim of recognizing an expressed attitude as positive or negative.

introduction

People speak in natural languages , i.e. languages that, unlike formal languages, do not convey meaning and information clearly and not only structurally and which are made more difficult to automatically process by computers. Computational linguistics explores how computers can still analyze natural language. For a long time one hoped for artificial intelligence , which tries to create intelligent systems, but since even modern computers are still a long way from this goal, the goals of language processing were narrowed down and turned to simpler but more promising methods. One such goal is to work out specific knowledge from texts, e.g. B. the topic or - as here - the attitude of the author to this topic. The field that deals with solving such tasks is called text mining , based on data mining , with which it has the basic ideas in common. The methods used by sentiment detection come from areas such as statistics , machine learning, and natural language processing .

Action

The task of sentiment detection is approached using statistical methods. In addition, the grammar of the utterances examined can be taken into account. Statistical analysis is based on a basic set of terms (or N-grams ) that are associated with positive or negative tendencies. The frequencies of positive and negative terms in the analyzed text are compared and determine the presumed attitude.

Based on this, machine learning algorithms can be applied. On the basis of preprocessed texts for which the attitudes are known, such algorithms can also learn which tendency they can be assigned to other terms.

With the help of natural language processing techniques , knowledge of natural language can be incorporated into the decision. For example, if the grammar of the texts is analyzed, machine-learned patterns can be applied to the structure.

literature

Bing Liu: Web Data Mining: Exploring Hyperlinks, Contents, and Usage Data , Berlin [u. a.], Springer, 2007.
Bo Pang et al. Lillian Lee: Opinion Mining and Sentiment Analysis , Hanover (MA), Now Publishers, 2008.
Haselmayer, Martin u. Marcelo Jenny (2017). Sentiment analysis of political communication: combining a dictionary approach with crowdcoding. Quality and Quantity 51 (6): 2623-2646.

swell

Unless otherwise stated, the article content is taken from the main source:

C. Ziegler: The measurement of opinion. Sentiment Detection: machine text understanding. In: iX - magazine for professional information technology . 10, Oct 2006, p. 106ff.