J
Jatin Sharma
Researcher at Microsoft
Publications - 10
Citations - 411
Jatin Sharma is an academic researcher from Microsoft. The author has contributed to research in topics: Transliteration & Latin script. The author has an hindex of 3, co-authored 9 publications receiving 361 citations. Previous affiliations of Jatin Sharma include LNM Institute of Information Technology.
Papers
More filters
Proceedings ArticleDOI
POS Tagging of English-Hindi Code-Mixed Social Media Content
TL;DR: The initial efforts to create a multi-level annotated corpus of Hindi-English codemixed text collated from Facebook forums are described, and language identification, back-transliteration, normalization and POS tagging of this data are explored.
Proceedings ArticleDOI
“I am borrowing ya mixing ?†An Analysis of English-Hindi Code Mixing in Facebook
TL;DR: The classification of Code-Mixed words based on frequency and linguistic typology underline the fact that while there are easily identifiable cases of borrowing and mixing at the two ends, a large majority of the words form a continuum in the middle, emphasizing the need to handle these at different levels for automatic processing of the data.
Proceedings Article
Query word labeling and Back Transliteration for Indian Languages: Shared task system description
TL;DR: This paper proposes a supervised approach of building a classier with monolingual samples together with a context-switching probability from Indian Language (IL) to English (Eng) and shows the best performing results.
Journal ArticleDOI
Deepfakes Classification of Faces Using Convolutional Neural Networks
TL;DR: A convolutional neural network-based model is proposed to detect the fake face images using transfer learning techniques from pretrained deep models such as VGG16 and ResNet50 to construct deep ensembles.
Two-Pass Adaptive Histogram based method for Restoration of Foggy Images
TL;DR: A Local Histogram Equalization (LHE) based method to improve the visual quality of foggy images by proposing to process blocks by overlapping them and using statistically optimal block size to obtain a good visual quality in reasonable time duration.