--- title: "Amir Zadeh — author profile and papers" description: "Amir Zadeh: 3 papers on Elamur Research (2017–2018), h-index 29 — affiliations, topics, co-authors and who cites them in our library." canonical: "https://research.elamur.ai/free/en/authors/amir-zadeh/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/amir-zadeh/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Amir Zadeh # Amir Zadeh Amir Zadeh works on multimodal sentiment and emotion analysis in video, where meaning is carried jointly by spoken words, facial expression and tone of voice. The papers here, from 2017 and 2018, ask how a model should combine these channels and how far it should rely on what surrounds a given utterance. The methods are neural: recurrent networks over per-utterance features drawn from text, audio and visual streams, with fusion and attention mechanisms built on top. The most cited of them, [Tensor Fusion Network for Multimodal Sentiment Analysis](https://md.research.elamur.ai/free/en/papers/zadeh-2017-tensor-fusion-network-multimodal-sentiment/) (2017), treats modeling within each modality and modeling across modalities as separate problems. It joins modality embeddings through an outer-product style operation so that pairwise and three-way interactions become explicit features. It also evaluates on splits that keep speakers apart between training and testing. [Context-Dependent Sentiment Analysis in User-Generated Videos](https://md.research.elamur.ai/free/en/papers/poria-2017-context-dependent-sentiment-analysis-user/) (2017) first encodes each utterance on its own for every modality. It then runs a contextual LSTM over the sequence of utterances in a video, so that neighbouring statements inform each sentiment label. The 2018 [Conversational Memory Network](https://md.research.elamur.ai/free/en/papers/hazarika-2018-conversational-memory-network-emotion-recognition/) moves from single-speaker opinion videos to two-person conversations and recognises emotion utterance by utterance. It keeps a separate recent history for each speaker, stores these as GRU-encoded memories, and uses attention to pick out the earlier turns that matter. In this way it models both the persistence of a speaker's own feelings and the pull of the conversation partner. Paper bylines connect Zadeh with [Carnegie Mellon University](https://md.research.elamur.ai/free/en/universities/carnegie-mellon-university/) in Pittsburgh from 2015 to 2024. Co-authors on several of the papers here include [Louis-Philippe Morency](https://md.research.elamur.ai/free/en/authors/louis-philippe-morency/), [Soujanya Poria](https://md.research.elamur.ai/free/en/authors/soujanya-poria/), Erik Cambria and [Devamanyu Hazarika](https://md.research.elamur.ai/free/en/authors/devamanyu-hazarika/). The work falls mainly under [Sentiment & emotion](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/), with links to dialogue systems and [Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/). Amir Zadeh has 3 papers in our library, in 2017–2018, most of them on [Sentiment & emotion](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/), [Dialogue systems](https://md.research.elamur.ai/free/en/topics/dialogue-systems/), [Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/); often with [Erik Cambria](https://md.research.elamur.ai/free/en/authors/erik-cambria/), [Louis-Philippe Morency](https://md.research.elamur.ai/free/en/authors/louis-philippe-morency/), [Soujanya Poria](https://md.research.elamur.ai/free/en/authors/soujanya-poria/). 3 papers in our library · 2017–2018 [OpenAlex](https://openalex.org/A5112033266) · [ORCID](https://orcid.org/0000-0002-5297-3571) · [Wikidata](https://www.wikidata.org/wiki/Q130959456) · [**h-index 29**](https://openalex.org/authors/A5112033266) · [61 works](https://openalex.org/works?filter=author.id:A5112033266) · [10,556 citations](https://openalex.org/works?filter=author.id:A5112033266&sort=cited_by_count:desc) OpenAlex, 2026-10-14 [3 works in our library](https://md.research.elamur.ai/free/en/authors/amir-zadeh/#papers) · [349 citations in our library](https://md.research.elamur.ai/free/en/authors/amir-zadeh/cited-by/) · 1 more work in our library · [⭐ Paper Star rank #706](https://md.research.elamur.ai/free/en/paper-stars/page/8/#star-706) The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Amir+Zadeh) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Amir+Zadeh) · [dblp](https://dblp.org/search?q=Amir+Zadeh) · [ACL Anthology](https://aclanthology.org/search/?q=Amir+Zadeh) ## Affiliations [🎓**Carnegie Mellon University** Pittsburgh · United States 2015–2024 · 38 papers](https://md.research.elamur.ai/free/en/universities/carnegie-mellon-university/) From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 3 more institutions on single bylines are not listed. ## Topics [Sentiment & emotion 3](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) · [Dialogue systems 1](https://md.research.elamur.ai/free/en/topics/dialogue-systems/) · [Vision & language 1](https://md.research.elamur.ai/free/en/topics/multimodal/) ## Works by year ### 2010s [**2018** 1 paper](https://md.research.elamur.ai/free/en/papers/hazarika-2018-conversational-memory-network-emotion-recognition/) · [**2017** 2 papers](https://md.research.elamur.ai/free/en/authors/amir-zadeh/year/2017/) ## Papers The most cited first (OpenAlex and our library). 1. **[Tensor Fusion Network for Multimodal Sentiment Analysis](https://md.research.elamur.ai/free/en/papers/zadeh-2017-tensor-fusion-network-multimodal-sentiment/)** *(2017 · 2,107 citations)* The paper addresses sentiment prediction for opinionated online videos, where spoken words, facial behavior and voice all shape the perceived sentiment. 2. **[Context-Dependent Sentiment Analysis in User-Generated Videos](https://md.research.elamur.ai/free/en/papers/poria-2017-context-dependent-sentiment-analysis-user/)** *(2017 · 1,019 citations)* Most multimodal sentiment systems for video label each utterance as if it stood alone and ignore what the speaker said just before and after. 3. **[Conversational Memory Network for Emotion Recognition in Dyadic Dialogue Videos](https://md.research.elamur.ai/free/en/papers/hazarika-2018-conversational-memory-network-emotion-recognition/)** *(2018 · 539 citations)* The paper tackles utterance-level emotion recognition in videos of two-person (dyadic) conversations. Earlier systems either ignored context or used LSTM-based context models, which summarize long histories poorly and weight all prior turns equally. Drawing on psychology, the authors model two forces: emotional… ## Also in our library 1 more work of theirs is in our library without a page here yet; it will appear as we publish them. ## Co-authors here 6 co-authors with a page, the most papers together first. - [Erik Cambria](https://md.research.elamur.ai/free/en/authors/erik-cambria/) ×3 - [Louis-Philippe Morency](https://md.research.elamur.ai/free/en/authors/louis-philippe-morency/) ×3 - [Soujanya Poria](https://md.research.elamur.ai/free/en/authors/soujanya-poria/) ×3 - [Devamanyu Hazarika](https://md.research.elamur.ai/free/en/authors/devamanyu-hazarika/) ×2 - [Navonil Majumder](https://md.research.elamur.ai/free/en/authors/navonil-majumder/) - [Roger Zimmermann](https://md.research.elamur.ai/free/en/authors/roger-zimmermann/) ## Cited by Papers of our library cite their works 349 times (of the 140,884 papers whose bibliographies we read); [18 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/amir-zadeh/cited-by/).