ITIrina Toma

Irina Toma

*This is an algorithmically generated author record based on publicly available information.

Irina Toma has a total of 8 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • INTELLIV – A Digital Solution for Old and Pre-modern Romanian Literature

    by Laurențiu-Marian Neagu, Irina Toma, Mihai Dascălu, Lucian Chișu

    Digitalization of old and pre-modern works is essential for preserving our literary identity and heritage. As such, INTELLIV expands on the previous INTELLIT project focused on modern Romanian literature. The web platform includes four components: the Chronology of Old Romanian Literature (DCLR), the Encyclopedia of Old Romanian Literature (ELIV), the Fundamental Texts of Old Romanian Literature (TLRV), and the Cartography of Old Romanian Literature (HLRV). Our methodology involves data preprocessing, indexing, and storage using Elasticsearch and a relational database for authentication, logging, and storing HLRV information. The platform ensures public access to data visualizations, whereas restricted editing access is provided to authenticated users. As such, the outcomes include the development of a web platform for data visualization and editing, as well as a chatbot to interact with the DCLR and ELIV corpora dynamically.

  • A question of trust: Business partnership networks before and during COVID-19 pandemic in Romania

    by Cezar Scarlat, Mihnea Andrei Călin, Denis Iorga, Teodor Neacşu, Irina Toma, Mihai Dascălu

    The impact of the COVID-19 pandemic on the economic environment had and continues to have long-lasting consequences. Also, the COVID-19 pandemic induced changes in business strategies and partnerships. As identifying new trustworthy organizational partners is crucial for enhancing the business resilience during the COVID-19 crisis, this chapter presents some results of the authors’ work in this respect: the aim is to investigate the business partnership networks before and during the COVID-19 pandemic. The authors opted for an opinion mining approach: they analyzed a pipeline of Romanian online news, between 2018 and 2022, using adapted Natural Language Processing (NLP) techniques and Social Network Analysis (SNA) metrics. As a result, their novel approach provides business leaders with a tool to compute an initial level of trust in a potential business partner as a centrality measure of the latter in business co-occurrence networks, automatically extracted from online news. A direct implication is this: the research results demonstrate that business press – when explored with deep learning means – can be considered a source of information needed to reasonably estimate the level of organizational business trust.