IntenCheck Blog

IntenCheck Blog

IntenCheck Sentiment Text Analysis & Big Data. What’s the difference?

IntenCheck Sentiment Text Analysis & Big Data. What’s the difference?

By Prof. Vladimir Shalack, Co-founder of Intentex and Sentiment Text Analysis Guru.

Let’s begin by looking at the Big Data and compare to what we do at IntentexBig Data analysis is the process of collecting, organizing and processing large sets of data (big data) to discover useful information.

For example, there is a story about a father who started complaining about receiving ads in the mail that were specifically targeted towards pregnant women. This had begun when advertising companies noticed that his daughter was exhibiting online behaviors of pregnant women, which were characterized by analyzing large amounts of data. This is how the family began to receive advertisements in the mail. Interestingly enough, the companies were correct about the pregnancy, even though the daughter wasn’t even aware of it at the time.

To the majority of people, such stories sound impressive and they become huge supporters of Big Data. However, looking from a scientific point of view, I think Big Data is degrading the science because this approach focuses on a blunt search for answers, instead of deep analysis, correct interpretation, and understanding of the matter.

In other words, earlier statistical studies took various random distributions and picked a small random sample to replicate the whole representation. Nowadays, no one looks at the distributions and instead takes the whole chunk of data and draws conclusions from that. Technically speaking, the results of Big Data analysis are mechanical sorting of whole chunks of data in order to obtain some information.

I do not want to suggest that this technique does not yield interesting results. However, same results can be obtained with equal success by using standard statistical methods based on a representative distribution. This is why I think all of the excitement around Big Data is just a marketing trick and nothing more.

The text analysis engine, which we have developed – IntenCheck is not related to Big Data. Rather, it is better to compare it to sentiment analysis, which aims to assess the emotions, attitude, communication style, tone, motivation and more.

Evaluating Emotions in IntenCheck is important because it is an assessment of subjective distortions in the perception of the surrounding world, and this can be used for many different purposes.

Apart from emotions, IntenCheck has a group of categories called Attitude (Semantic Differentials), which also shows subjective relations to the objects and phenomenon of the surrounding world. This is based on the work of Charles E. Osgood. It turned out that regardless of the emotions we distribute to all objects, we really see things as a part of a three-dimensional space specified by the axes of PositiveNegative, StrongWeak, ActivePassive. Comparing the evaluations of semantic differentials of two people, we can get an idea about the differences between the two attitudes towards the surrounding world as well as track the dynamics of any changes in attitudes.

In addition, we are also focused on finding out the hidden internal psychological characteristics of  the people. For this assessment, we use the representational systems idea, which originates from Neuro-Linguistic Programming (NLP) in Communication Style analysis category. These categories are more stable and allow to evaluate the effectiveness of communication.

The Timeline category helps us understand how people perceive and are oriented in time. Less successful people like to talk about what happened in the past. Whereas, more successful people, although not forgetting the past, focus on the future. This is because change is only possible in the future, one cannot return and change the past, and correct mistakes that already happened.

The measurements in the Motivation category are related to the directions of motivations. People are usually motivated either by moving Away from something that they don’t desire or by moving Towards something that they want. This is an area of study of NLP.

Another very important group of categories when evaluating the communication process is Perceptual Positions. If for example, one talks about himself/herself and his/her own problems, it is unlikely that the other person will be interested in such a conversation. In communication, it is important to understand how a person communicates either from his own viewpoint or from other perceptual positions. When you know this, you can match his point of view on reality and interact in a much more effective way. This is also an area of NLP.

Therefore, most of the IntenCheck results not only capture the statistical data of a particular statement but go further to identify the deeper psychological characteristics of the author. In the future, I hope, the system will be supplemented by more categories.

To summarize, our system in its full capacity will be able to effectively evaluate any communication and create prognostic reports. For example, if you receive an angry emotional letter from someone you know your first response might also be highly emotional, which may destroy the relationship. However, in order to establish a long lasting communication with that person, it might be better not to be guided by emotions manifested from reading the letter in your response but rather based on more stable characteristics like a representative system, semantic differentials, motivations, time and position.