Cross-Topic Analysis

The text mining techniques allows to capture and understand people thoughts, opinions and preferences on social networks and web with respect to a certain discussed topic.

In particular, by means of the Cross-Topic Analysis it is possible to understand how different analysis relate one each other. This is performed through the computation of joint frequency tables and conditional probability tables.

Project Details

Category: Cross-Topic
Sector: Fashion
Period of analysis: 3 Months
Total analysed mentions: 800
Sources: Facebook

For instance, the cross-topic analysis allows to understand how each trending topic is related to the sentiment. The Sentiment X Topic table represents the percentange of occurence (or joint frequency) of each analysed topic with respect to the three level of sentiment (positive, neutral and negative).

It is also possible to estimate the probability of a document or post of being negative, neutral or positive given the fact that it speaks about a certain topic. This is done by means of the conditional probability tables, or using the Polar Plot.

In this analysis the polar plot shows immediately that posts related to Quality or Assistance have higher probability of being negative, while mentions talking about Bags are more likely positive.

The cross-topic analysis is not limited to “cross” trending topics with sentiment. It allows to combine together any analysis you are interested in.

For instance, it could be valuable to see the interaction between the topics and the level of satisfaction, that is to study the relationship between the topics discovery and the satisfaction analysis. The Topic X Satisfaction and the Satisfaction | Topic tables allow exactly to study their joint frequencies and their conditional probabilities.


Recent Analysis