Publications

Research & Publications

My academic contributions to AI/ML research, including peer-reviewed publications and ongoing thesis work.

Peer-Reviewed Publications

FeaturedA* Ranked
July 2025

Translative Neural Team Recommendation: From Multilabel Classification to Sequence Prediction

Kap Thang, Hawre Hosseini, Hossein Fani

SIGIR '25 - Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval

We reformulated team recommendation as a sequence-to-sequence task using transformer architectures, achieving up to 82x improvement over existing feedforward neural approaches. Our method was evaluated on 4 large-scale datasets (DBLP, USPTO, IMDB, GitHub) with distinct skill/expert distributions, consistently outperforming baselines across all metrics.

Team RecommendationTransformersSequence-to-SequenceNeural Networks
View PaperCode

Thesis

MSc ThesisIn Progress
Expected: April 2026

CERA: Context-Engineered Reviews Architecture for Synthetic ABSA Dataset Generation

University of Windsor • Supervised by Dr. Luis Rueda

A three-phase pipeline for generating high-quality synthetic review datasets using LLMs. Key innovations include SIL (Subject Intelligence Layer) for factual grounding via web search, MAV (Multi-Agent Verification) for reducing hallucination through majority voting, and MDQA (Multi-Dimensional Quality Assessment) for comprehensive evaluation. Preliminary results show synthetic data achieving up to 93.2% of real human-annotated dataset performance.

LLMsSynthetic DataNLPAspect-Based Sentiment AnalysisData Generation