ICT4Growth is a research project that PaloServices undertook to develop algorithmic software that fuel its innovative platform. More specifically the R&D areas which PaloServices targeted and implemented were the following:
Automatic Resource Discovery:
an optimized process that discovers new sources related to entities like companies and organizations. This top layer process provides data that can be analyzed from other processes using Natural Language Processing techniques.
Named Entity Recognition:
disambiguation of “real entities” found in text excerpts vs. references unrelated to a specific entity (such as a company, person, location etc.) enriches the semantic knowledge of any text content. This process aims at providing meaningful information in relation to a text excerpts, not only by tagging such recognized entities, but also by grouping multiple forms of an entity to the same entity name.
the extraction of a positive, negative or neutral sentiment or opinion is another step forward to understanding the real content of a text excerpt. Sentiment Analysis can be applied to entities extracted by NER and can provide with information such as the overall reputation of an organization or individual, or the impact of a specific event related to an entity.
by first clustering the articles based on content, this process is able to extract a small but comprehensive summary of a group of articles that refer to the same topic. This offers a very quick but semantically correct view of text that may spread to many paragraphs.
is able to discover multiple aspects that are present in text like news and social media. Aspect mining can be applied on recognized entities to reveal insights in the form of aspects or “views” relating to a specific entity.
The aforementioned algorithms along with many other technologies and the supporting infrastructure serve to the core of Palo’s services and offer a wide range of facilities, ranging from innovative news reading applications to specialized brand reputation services.