Thanks to technological advances, storage, exploitation and information extraction of ever increasing amounts of data becomes more important every day. As a result, it becomes imperative to utilize data, develop smart decision supports systems and apply such capabilities to all sectors using Artificial Intelligence, which all authorities agree to be the technology that will determine and shape the future.
The Speech and Language Processing Systems Division of TÜBİTAK BİLGEM BTE possesses significant knowledge in the field of Artificial Intelligence thanks to its expert staff and conducting successful government projects for many years. The Text Analytics product developed within the Division provides solutions with performance and flexibility to meet the needs of both the public and the private sector.
The main operation in text analytics is automatic extraction and interpretation of information in documents or texts (covering all kinds of structured and unstructured text such as news, short or long texts, product reviews, social media messages). This makes it possible to utilize information, which emerges as one of the most fundamental needs today due to advancing technology and increasing amounts of data.
▪ Application of Artificial Intelligence and Natural Language Processing techniques,
▪ Dynamic learning models and high performance rates,
▪ Text preprocessing for Turkish and correction of nonstandard text (normalization)
▪ Operation on all kinds of short or long texts,
▪ Ability to define desired number of categories and classes,
▪ Ability to perform hierarchical classification,
▪ Automatic extraction of named entities (person, location, organization, other, etc.)
▪ Spotting certain terms according to usage (weapon serial number, identity number, gps coordinates, license plate numbers etc.),
▪ Automatic categorization/classification of texts,
▪ Prioritization and scoring of results.