Devamanyu Hazarika

Ph.D. Student,
National University of Singapore


I am a second-year Computer Science PhD student at National University of Singapore, advised by Prof. Roger Zimmermann.   My research interests primarily lie in deep learning for natural language processing and multimodal affective computing.

I completed by UG major in Computer Science from NIT Warangal, India where I also received the Institute Gold Medal for best academic performance in the undergraduate course.



 Self-Attentive Feature-level Fusion for Multimodal Emotion Detection  
Devamanyu Hazarika, S Gorantla, S Poria, R Zimmermann.
MR2AMC, IEEE MIPR, Miami, Florida (2018).    New !   

Multimodal emotion recognition is the task of detecting emotions present in user-generated multimedia content. Such resources contain complementary information in multiple modalities. A stiff challenge often faced is the complexity associated with feature-level fusion of these heterogeneous modes. In this paper, we propose a new feature-level fusion method based on self-attention mechanism. We also compare it with traditional fusion methods such as concatenation, outer-product, etc. Analyzed using textual and speech (audio) modalities, our results suggest that the proposed fusion method outperforms others in the context of utterance-level emotion recognition in videos.
Uni-dimensional Self-attentive Fusion.
Multi-dimensional Self-attentive Fusion.

Devamanyu Hazarika, R Bajpai, K Singh, S Gorantla, E Cambria, R Zimmermann.  Aspect-Sentiment Embeddings for Company Profiling and Employee Opinion Mining   In: CICLing 2018, Hanoi, Vietnam.    New !   

With the multitude of companies and organizations abound today, ranking them and choosing one out of the many is a difficult and cumbersome task. Although there are many available metrics that rank companies, there is an inherent need for a generalized metric that takes into account the different aspects that constitute employee opinions of the companies. In this work, we aim to overcome the aforementioned problem by generating aspect-sentiment based embedding for the companies by looking into reliable employee reviews of them. We created a comprehensive dataset of company reviews from the famous website and employed a novel ensemble approach to perform aspect-level sentiment analysis. Although a relevant amount of work has been done on reviews centered on subjects like movies, music, etc., this work is the first of its kind. We also provide several insights from the collated embeddings, thus helping users gain a better understanding of their options as well as select companies using customized preferences.
Projection of the Aspect-sentiment Embeddings of the companies. Note: The same color represents companies from the same sector.

• E Cambria, S Poria, Devamanyu Hazarika, K Kwok.  SenticNet 5: Discovering conceptual primitives for sentiment analysis by means of context embeddings   In: AAAI (2018), New Orleans, Louisiana, USA.   

With the recent development of deep learning, research in AI has gained new vigor and prominence. While machine learning has succeeded in revitalizing many research fields, such as computer vision, speech recognition, and medical diagnosis, we are yet to witness impressive progress in natural language understanding. One of the reasons behind this unmatched expectation is that, while a bottom-up approach is feasible for pattern recognition, reasoning and understanding often require a top-down approach. In this work, we couple sub-symbolic and symbolic AI to automatically discover conceptual primitives from text and link them to commonsense concepts and named entities in a new three-level knowledge representation for sentiment analysis. In particular, we employ recurrent neural networks to infer primitives by lexical substitution and use them for grounding common and commonsense knowledge by means of multi-dimensional scaling.
A sketch of SenticNet~5's graph showing part of the semantic network for a primitive INTACT.


• S Poria, E Cambria, Devamanyu Hazarika, N Mazumder, A Zadeh, L Morency.  Multi-level multiple attentions for context-aware multimodal sentiment analysis.   In: ICDM, New Orleans (2017).

• S Poria, E Cambria, Devamanyu Hazarika , N Mazumder, A Zadeh, L Morency.  Context-dependent sentiment analysis in user-generated videos.   In: ACL, pp. 873-883, Vancouver (2017).

• A Singh, Devamanyu Hazarika, A Bhattacharya.  Texture and Structure Incorporated ScatterNet Hybrid Deep Learning Network (TS-SHDL) For Brain Matter Segmentation.   IEEE International Conference on Computer Vision Workshops (ICCVW) 2017.

• E Cambria, Devamanyu Hazarika, S Poria, A Hussain, RBV Subramaanyam.  Benchmarking Multimodal Sentiment Analysis.   18th International Conference on Computational Linguistics and Intelligent Text Processing (CICling), Budapest, 2017.


• S Poria, E Cambria, Devamanyu Hazarika, P Vij.  A deeper look into sarcastic Tweets using deep convolutional neural networks.   26th International Conference on Computational Linguistics (COLING), Osaka (2016).



• T Young, Devamanyu Hazarika , S Poria, E Cambria.  Recent trends in deep learning based natural language processing.   Under review at IEEE CI Magazine.   

R3: Recommended Reading Resources

• Stable computations for softmax, sigmoid and its application in Logistic Regression : the LogSumExp trick.