Mohammad is a data scientist who loves solving challenging real-world problems using big data and machine learning. His research interests span mainly in data mining and machine learning using large-scale datasets, with an emphasis on their applications in health informatics and social informatics. His research has been published in several major academic venues, including SIGIR, WSDM, ICMR, etc. He is currently a research fellow Prof. Rumi Chunara working on wellness profiling of users and communities on social networks. He obtained a PhD in Integrative Sciences and Engineering from the National University of Singapore, advised by Prof. Tat-Seng Chua.
One paper on "Vertical Domain Text Classification: Towards Understanding IT Tickets using Deep Neural Networks ", was accepted by AAAI 2018.
One paper on "360° user profiling: past, future, and applications", was published by ACM SIGWEB Newsletter.
Our paper on "On the Organization and Retrieval of Health QA Records for Community-based Health Services", was selected for Best Paper Award by IJCAI, BOOM.
My area of interest spans Media Search and Retrieval. Within the range, my current research focus is on the area of Knowledge Management and Organization in Social Media, KnowledgeGraph extraction and embedding, and Multimodal information fusion and rerrieval. Specifically, I am interested in structuralizing and organizing User Generated Contents (UGCs) for healthcare and well-being domain.
Knowledge Organization in Health Domain
Community-based services leverage the wisdom of crowd through supporting communication, information sharing, and collaboration between individual users. Unsurprisingly, the impact of social media has been extended to the health care domain, as consumers have begun to seek information, share knowledge and experiences online. Therefore it imposes a greater impact on people’s daily information seeking, knowledge construction, and decision making. Although online community-based health services (CBHS) accumulate a huge amount of knowledge and grow at a continuously increasing pace, this knowledge is not effectively accessible due to the unstructured, noisy and opaque nature of data. To enhance the aggregation, navigation, and access into knowledge of the crowd, this research explores techniques to automatically analyze, organize, and retrieve large scale user generated contents (UGCs) on CBHSs.
Multi-Modal Information Retrieval and Ranking
Information reranking is to recover the true order of the inittial search results. Traditional reranking approaches, such as graph-based and pseudo-based, have achieved great success for uni-modal queries. They, however, suffer from some intrinsic limitations. (1) They only capture the pairwise relations instead of high-order relations, which may lead to information loss; (2) They also usually simply concatenate heterogeneous features into one vector that may cause the curse of dimensionality, and neglect the effects of different type of features. In this work, we investigate to find a unified multi-modal framework for multi-modality information retrieval system, where the queries can be mixture of texts, images, videos and audios.
Please send e-mail to me to request (p)reprints of papers that do not have a downloadable pdf associated with them.
- Aleksander Farseev, Mohammad Akbari, Ivan Samborskii, Tat-Seng Chua, 360° user profiling: past, future, and applications, ACM SIGWEB Newsletter.
- Mohammad Akbari, Xia Hu, Liqiang Nie, Tat-Seng Chua, On the Organization and Retrieval of Health QA Records for Community-based Health Services, IJCAI, BOOM Workshop, Best Paper Award.
- Mohammad Akbari, Xia Hu, Liqiang Nie, Tat-Seng Chua, From Tweets to Wellness: Wellness Event Detection from Twitter Streams, AAAI Conference on Artificial Intelligence. [Source code and Data Soon!].
- Mohammad Akbari, Liqiang Nie, and Tat-Seng Chua, aMM: Towards adaptive ranking of multi-modal documents, International Journal of Multimedia Information Retrieval.
- Song, Xuemeng, Liqiang Nie, Luming Zhang, Mohammad Akbari, and Tat-Seng Chua, Multiple social network learning and its application in volunteerism tendency prediction, Annual ACM SIGIR Conference.
- Farseev, Aleksandr, Liqiang Nie, Mohammad Akbari, and Tat-Seng Chua, Harvesting multiple sources for user profile learning: a big data study, International Conference on Multimedia Information Retrieval.
- Nie, Liqiang, Yi-Liang Zhao, Mohammad Akbari, Jialie Shen, and Tat-Seng Chua, Bridging the vocabulary gap between health seekers and healthcare knowledge, IEEE Transaction on Knowledge and Data Engineering.
- Nie, Liqiang, Mohammad Akbari, Tao Li, and Tat-Seng Chua, A joint local-global approach for medical terminology assignment, In Medical Information Retrieval Workshop at SIGIR 2014.
- Nie, Liqiang, Tao Li, Mohammad Akbari, Jialie Shen, and Tat-Seng Chua, Wenzher: Comprehensive vertical search for healthcare domain, Annual ACM SIGIR Conference.
Teaching Assistant, National University of Singaopore, 2015
Social Media Computing, CS4242
Lecturer, Azad University, 2006-2015
Neural Network, Artificial Intelligence, Compiler Design, Programming Languages Design and Implementation
Invited Lecturer, Iran University of Science and Technology, 2008-2009
Compiler Design, Artificial Intelligence, Advanced Programming
Journal and Conference Reviewer
- IEEE Transaction of Knowledge and Data Engineering (TKDE).
- ACM Transaction on Knowledge Discovery from Data.
- Pattern Recognition Letters Journal.
- World Applied Science Journal
- ACM Multimedia Conference ACM MM 2015
- World Wide Web and Population Search at AAAI 2015
- IEEE International Conference on Healthcare Informatics 2015 (ICHI 2015)
- IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining ASONAM 2015
- IEEE AINL-ISMW 2015
I I occasionally update my blog here.