SentiCognitiveServices - the new generation of marketing automation
The summary of the project co-founded with EU grant:
Social media analysis software and services make the market that was worth $1.6 billion in 2015 and is said to raise up to $5.4 in 2020. Nowadays, online monitoring solutions provide users with simple mention analysis that is based on word count or sentiment analysis using text corpora and AI techniques. As a result, getting the answer to “what do the internauts really think of my brand?”, asked by marketing, client service, and product management teams, needs manual mention examination made by analytics teams to create dedicated qualitative reports. Given the mention scope reaches millions every day, the proper summary of online discussions about a chosen topic needs a huge workload of dedicated analytics specialists.
The goal of this project is to develop a technology allowing full automation in creating reports. What needs to be done to provide such technology:
- Natural language processing research with the emphasis on social media communication.
- Extraction of data from online mentions with rule-based machine learning systems.
- Inference based on multiple information extraction models such as aspects, objects, valuable quotes, product comparisons.
Next, as a part of project development, there are to be built application modules which will monitor dynamically generated sites, create automatic reports and detect online crises.
The outcome of project activities will result with implementation of new features and improvement of current SentiOne tool. The demand for which has been confirmed by market research and based on the letters of intent from potential clients. Therefore, the platform creators will come by cutting-edge product offering functionalities without equal on a global scale.
Project value: PLN 11 699 568.70
Co-financing: PLN 8 682 464.57
- number of project: POIR.01.01.01-00-0806/16
- duration: 07/2017 - 06/2020
- measure: 1.1 R&D Projects of Enterprises
- sub-measure: 1.1.1 Industrial Research and Development Work Conducted by Enterprises
Project is supported by ERDF