Vartani Spellcheck -- Automatic Context-Sensitive Spelling Correction of OCR-generated Hindi Text Using BERT and Levenshtein Distance

no code implementations • 14 Dec 2020 • Aditya Pal, Abhijit Mustafi

Automatic spelling error detection and context-sensitive error correction can be used to improve accuracy by post-processing the text generated by these OCR systems.

named-entity-recognition Named Entity Recognition +4

Anubhuti -- An annotated dataset for emotional analysis of Bengali short stories

no code implementations • 6 Oct 2020 • Aditya Pal, Bhaskar Karn

We explain the data collection methods, the manual annotation process and the resulting high inter-annotator agreement of the dataset due to the linguistic expertise of the annotators and the clear methodology of labelling followed.

Emotion Classification feature selection

PinnerSage: Multi-Modal User Embedding Framework for Recommendations at Pinterest

no code implementations • 7 Jul 2020 • Aditya Pal, Chantat Eksombatchai, Yitong Zhou, Bo Zhao, Charles Rosenberg, Jure Leskovec

Latent user representations are widely adopted in the tech industry for powering personalized recommender systems.

Clustering Recommendation Systems

Hierarchical Temporal Convolutional Networks for Dynamic Recommender Systems

no code implementations • 8 Apr 2019 • Jiaxuan You, Yichen Wang, Aditya Pal, Pong Eksombatchai, Chuck Rosenberg, Jure Leskovec

Recommender systems that can learn from cross-session data to dynamically predict the next item a user will choose are crucial for online platforms.

Session-Based Recommendations