|Target Users||B2B Publishing firms, bookstores, libraries|
|Nature of the initiative||Public|
Founder & CEO
Tekstum is the first technology to provide real time scientific analysis of qualitative information (the opinions and feelings that readers directly express) specifically for the book retail, distribution and publishing industries. With a state-of-the-art algorithm built with Big Data and Artificial Intelligence, Tekstum technology discovers the emotions that a book transmits to its readers.
The startup offers its subscribers APIs that identify, measure and analyze emotions from readers in order to enrich webs of retailers and publishing houses, improve recommendation results, and strengthen the success ratio of different stakeholders (publishers, authors, marketing executives, etc.). The API is a value-added service that provides a more personalized experience for customers and a more insightful analysis for publishers, thus benefiting clients from both ends of the spectrum.
Founded in 2014 by Marc Santandreu, former publisher, Tekstum revolutionizes how publishers market and how customers search for and buy books through an innovative technology that synthesizes, analyzes and visualizes the real time emotions and experiences of readers.
Every year the number of new book titles in the market grows exponentially, generating large amounts of data at fast rates. Tekstum breaks this wall of content and its limitations by not being based on past sales, searches, or the inaccurate self-evaluation star system but on the scientific analysis of the opinions and feelings that readers are expressing.
The Tekstum engine uses natural language processing techniques to find patterns in qualitative data comprised of the opinions and feelings of actual readers in order to draw conclusions regarding the sentiment and emotions of how they experience a book.
Tekstum has built an innovative technology that facilitates discovery of new titles, better understanding of markets and distinguishes the book shopping experience.
Currently Tekstum has three main functions: 1) to generate reports based on qualitative information to analyze the overall sentiment trends of readers in specific markets; 2) to enrich websites that sell books by visualizing the results of a sentiment analysis based on actual opinions of readers; 3) to power a novel recommendation system based on reader emotions and sentiment rather than past purchases or searches, such as more traditional recommendation algorithms.
The role of technology
Tekstum combines artificial intelligence and big data to build an intelligent algorithm that uses natural language processing techniques to find patterns hidden in large volumes of qualitative data. It analyzes this rich domain-tuned information to uncover insights regarding specific demographics and markets, as well as, and most importantly, the emotions of real readers.
Tekstum performs Sentiment Analysis of tweets and reviews written by readers of books. Sentiment analysis is a scientific study that permits the automatic identification of the opinions (either positive, negative, neutral, and in some cases null) of the author of a text. Tekstum uses this analysis to automatically detect the polarity of a Review or a Tweet of a specific book. With this information, we are then able to calculate the global sentiment of readers towards a given book.
Both for tweets and reviews, what makes Tekstum's Sentiment Analysis tool special and novel with respect to work done by other companies lies in one key aspect: Tekstum focuses on a specific domain, that of books and the publishing sector, which is a domain that hasn't been considered much until now; focusing on a specific domain allows Tekstum to achieve much higher accuracy for the tool.
As of February 2016, Tekstum has begun to commercialize its product. Until now, the focus of Tekstum has been on the Spanish market, more specifically, the Catalonia region. With more than 3,000 users, Spanish publishers of all sizes (included big firms) have energetically embraced the Tekstum technology and its possibilities to shape their marketing strategies and campaigns. In addition the company is excited to launch the English version of its technology at the end of this year.
Furthermore, as Tekstum has established itself as a key reference resource for the sector. Currently, Tekstum is participating in the Google Startup Program, the startup is co-financed by the Catalan government and it has also been shortlisted for the Quantum Innovation Award of the London Book Fair in 2016.