In addition to sentiment analysis, which I will soon discuss, natural language processing has many applications today.
Emotion analysis on the Internet and opinion mining, some examples
At this point, we have everything we need to analyze emotions on the Internet.
Big Data, natural language processing and now, indonesian phone numbers opinion mining, which is the application of these two elements (plus data mining) to identify and extract information from what users say.Subjective information, of course.
This sentiment analysis focuses on these tasks:
Analyze polarity
Find out whether comments about a brand are positive, negative or neutral.
Analyze the characteristics of a product
Can you imagine knowing what users think about a product?
The simplest algorithms identify only whether a comment is positive or negative, but advanced ones are able to detect subtleties.
1. Entity detection
For example, the company Bitext has a tool capable of detecting entities in a text, that is, it isolates them and is able to know what each entity is.
Bitext example
In this way, they can analyze and extract concepts (such as having good after-sales service) or classify feelings into positive and negative.
2. Sensing Senses
Sentiment analysis example
In this example from the company Bitext, you can see how positive, negative and neutral feelings are located.
3. Polarization of opinions
Another good example of polarization of opinions is the opinion aroused by the death of the Iron Lady, Margaret Thatcher, in the United Kingdom.
Polarization of opinions
The research was conducted through Twitter with an analysis of more than 200,000 tweets by the company Pulsar Platform.
This was the result.
4. Product Feature Reviews
Another interesting company is GeomIndex , a benchmark in the automotive sector.
With its tools you can find out the opinions of users in real time and the reputation they have on the Internet.
In this study, they analyzed 354,154 opinions on 44 different brands, taking into account 41 parameters.
All of these applications help reveal what customers feel and what they want and what they associate with what.
5. Brand and values association
TNS is also working in this direction, but it does so by analyzing advertising and sponsorships in real time.
In this example, you can see which brands are most associated with Rafa Nadal.