Startups continue to talk about; and this time, it is the turn of the people of Tekstum who have opted for the launch of a platform based on Big Data and Natural Language Processing (NLP), capable of analyzing the feelings and emotions that the books transmit to their readers to transform them into useful data for the publishing industry.
Spokesmen of the firm have explained that the tool analyzes the information that appears on the Internet and identifies keywords, expressions and styles of the works, establishing a valuation, the Tekstum Score, based on the parameters studied: Sentiment Analysis of the comments, Social Engagement of the book and social reputation of the author.
The publishing world uses sales figures as a basis to assess whether or not a book has been liked by the public, but often this data is insufficient to know what users want to read.
Regarding the benefits offered by this proposal, the informants have mentioned that for the publishers, Tekstum allows to reduce the uncertainty before each launch, to focus more the model on the demand, to know better the profile of the possible readers and to discover new values national or international; while for readers, Tekstum becomes a space to find recommendations made by other readers and processed by the algorithm.
Currently, the startup has launched the Beta version through its website www.tekstum.com. Users who access your website will be able to find a selection of the books best rated by the Tekstum algorithm and know the global feeling that they transmit in addition to the reader profile for each work, the informants have pointed out, adding that during the beta phase, Tekstum recommend Similar books for each title analyzed and offer to all those who register the possibility to request a free analysis of the book you want.
Tekstum, which is part of the incubator specialized in Big Data Incubio, is an initiative of Marc Santandreu and Marc Martnez, both economists with experience in the publishing world () The startup team is also integrated by a data scientist, a computer engineer And a linguist.
I like this: