This is the 15th edition of the Forum for Information Retrieval Evaluation (FIRE 2023). Continuing the tradition of the last 14 years, the evaluation tracks at FIRE have expanded to encompass new domains and new languages. With nine evaluation tracks and participation from over 50 teams, this is one of the biggest FIRE workshops to date. This year FIRE realized an interesting mix of tasks from IR/NLP viz. Information Retrieval in Software Engineering (IRSE), Cross-Lingual Information Retrieval for African Languages (CIRAL), sentiment analysis, summarization on different kinds of datasets like social media, legal text etc. in different languages including English, Dravidian, etc. Hate Speech and Offensive Content Identification in Indo-European Languages (HASOC), focused on identifying various categories of hate speech from social media posts. Indian Language Summarization (ILSUM) offered a test bed for summarization in Hindi, Gujarati and English. Information Retrieval in Software Engineering (IRSE) offered a task on classification of source code. These tracks were continued from last year with similar to slightly modified problem statements. FIRE 2023 saw the induction of many new and interesting tracks. Cross-Lingual Information Retrieval for African Languages (CIRAL) was dedicated to the retrieval between English and four African languages which are Hausa, Somali, Swahili and Yoruba. This is the first time that FIRE has included African languages in its scope since its inception. Uncovering Truth in Social Media through Claim Detection and Identification of Claim Spans (CLAIMSCAN) addressed detection of a claim and it's span on social media text. Machine Translation for Indian Languages (MTIL) offered a translation setup between a vast number of Indian language pairs. They also touched upon important domains like Heathcare and Governance. CoLI-Tunglish: Word-level Language Identification in Code-mixed Tulu Texts task was dedicated on Tulu language identification in Code-mixed Tulu texts. Scarcasm Identification of Dravidian Languages (Malayalam and Tamil) in DravidianCodeMix 2023 (DravidianCodeMix) provided a task to identify sarcasm and sentiment polarity of the code-mixed dataset of comments/posts in Tamil-English and Malayalam-English collected from social media. The tracks, in general, received good and uniform participation from many research groups. Each track had a dedicated Program Committee (PC member names and affiliations are as below). We express our gratitude to all the track organizers for their efforts in organizing these tracks, as well as to the participants for their enthusiastic participation. A special thanks to our sponsor 'ACM SIGIR Special Interest Group on Information Retrieval' for the continued support.
Trackwise PC members Artificial Intelligence on Social Media (AISoMe)
Cross-Lingual Information Retrieval for African Languages (CIRAL)
Uncovering Truth in Social Media through Claim Detection and Identification of Claim Spans (CLAIMSCAN)
Hate Speech and Offensive Content Identification in English and Indo-Aryan Languages (HASOC) Information Retrieval in Software Engineering (IRSE) Machine Translation for Indian Languages (MTIL) Scarcasm Identification of Dravidian Languages (Malayalam and Tamil) in DravidianCodeMix 2023 (DravidianCodeMix)
Dr. Subalalitha Chinnaudayar Navaneethakrishnan, Department Of Computer Science & Engineering, SRM Institute Of Science And Technology, Tamil Nadu, India
Word-level Language Identification in Code-mixed Tulu Texts (CoLI-Tunglish) Indian Language Summarization (ILSUM)