Services and solutions for text analysis and knowledge extraction.

Talia is a spin-off company of CNR ILC (‘A. Zampolli’ Institute of Computational Linguistics) and GruppoMeta.

It develops and delivers tools for text processing and knowledge extraction based on Automatic Language Processing and Artificial Intelligence techniques.

It offers specialised solutions and services in morphological analysis, syntactic analysis, name entity recognition, meaningful knowledge extraction, classification, clustering, ontology learning, ontology population and sentiment analysis in Saas (software as a service) mode.

Topic

Talia develops and delivers tools for text processing and knowledge extraction based on Automatic Language Processing and Artificial Intelligence techniques.

It offers specialised solutions and services in morphological analysis, syntactic analysis, name entity recognition, meaningful knowledge extraction, classification, clustering, ontology learning, ontology population and sentiment analysis in Saas (software as a service) mode.

  • Hybrid approach. It adopts a hybrid approach based on natural language processing (NLP) and artificial intelligence (AI) techniques.
  • Specialised services. It offers specialised services for creating:
    • annotated corpuses
    • configurations of ontologies
    • specialised systems for entity recognition
    • clustering systems and tools to support QA tools, chat bots, etc.
  • SaaS. The tools available on the TALIA.CLOUD enable applications of all kinds to be enriched with special functions:
    • text editor
    • e-learning platforms
    • CRM tools
    • e-commerce platforms
    • cataloguing and document management software
    • Chat bots
    • recommendation systems
    • supply and demand matching
    • automatic metadata extraction and indexing
    • classification and collation of documents
    • knowledge discovery systems

TALIA.CLOUD will make use of open source solutions implemented by CNR, such as the T2K platform and other existing tools implemented by research groups and foundations.

Services and solutions

Talia solutions are delivered with a SaaS (software as a service) approach to maximise efficiency, remote access to services, scalability and security, and optimise investments.

  • Tools. Talia provides tools for:
    • Morphological-syntactic analysis of texts
    • Extraction of relevant nominal entities such as names, dates, places, institutions, age
    • Extraction of meaningful knowledge and relationships between concepts
    • Automatic creation of conceptual maps
    • Implementation of QA tools over vertical domains
    • Text generation and classification
    • Sentiment analysis
    • Dialogue generation
    • Document clustering
    • Proposal of related content
  • Innovative services. TALIA provides services that are highly innovative compared to the state of the art in the field, in terms of the ability to analyse the text from diverse perspectives, to combine different dimensions of analysis, and to generate knowledge that would otherwise be impossible to extract:
    • Training models for AI algorithms over vertical domains
    • Creation and extraction of specialised ontologies
    • Annotation of text corpora
    • Tuning and customisation of chat bots, QA systems and recommender engines
  • Extraction of the writer’s opinions. Identification and extraction of the writer’s opinions in relation to the extracted information content.
    By analysing the style used by the writer, TALIA services are able to identify: the age of the writer, their level of schooling, gender and mother tongue – if different from the language in which the text is written – and even whether a given portion of the text was written by a particular person or not..
    Text analysis of different genres at various levels of accuracy:
    • journalistic, literary, domain-specific texts (legal, technical, scientific, biomedical)
    • texts drawn from social media, such as Twitter, Facebook or review sites, or texts produced by students using e-learning platforms.