{"id":196,"date":"2021-08-02T16:09:19","date_gmt":"2021-08-02T21:09:19","guid":{"rendered":"https:\/\/groups.cs.umass.edu\/zamani\/?page_id=196"},"modified":"2021-08-02T16:09:19","modified_gmt":"2021-08-02T21:09:19","slug":"research-topics","status":"publish","type":"page","link":"https:\/\/groups.cs.umass.edu\/zamani\/research-topics\/","title":{"rendered":"Research Topics"},"content":{"rendered":"\n<p>My research interests include various aspects of information retrieval, recommender systems, and text mining. My major contributions and current research focus include:<\/p>\n<p><strong>Neural Information Retrieval<\/strong>: Following our initial study on distributed representations for information retrieval (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1225\" target=\"\\_blank\" rel=\"noopener\">ICTIR16a<\/a>,\u00a0<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1234\" target=\"\\_blank\" rel=\"noopener\">ICTIR16b<\/a>,\u00a0<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1224\" target=\"\\_blank\" rel=\"noopener\">CIKM16<\/a>), we have developed multiple neural network models for the fundamental IR tasks. They include relevance-based word embedding (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1267\" target=\"\\_blank\" rel=\"noopener\">SIGIR17a<\/a>), context-aware query representation (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1268\" target=\"\\_blank\" rel=\"noopener\">WWW17<\/a>), and semi-structured document representation (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1290\" target=\"\\_blank\" rel=\"noopener\">WSDM18<\/a>). We have also invented the first learning to rank model that can retrieve (as opposed to re-rank) documents from large-scale collections, called standalone neural ranking model (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1302\" target=\"\\_blank\" rel=\"noopener\">CIKM18a<\/a>).<\/p>\n<p><strong>Weak Supervision for IR<\/strong>: We have developed a student-teacher learning model for training machine learning IR models with no labeled data, called\u00a0<em>weak supervision<\/em>. We have empirically shown the effectiveness of weak supervision in various IR tasks, including ad-hoc retrieval (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1266\" target=\"\\_blank\" rel=\"noopener\">SIGIR17b<\/a>), representation learning (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1267\" target=\"\\_blank\" rel=\"noopener\">SIGIR17a<\/a>), and query performance prediction (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1300\" target=\"\\_blank\" rel=\"noopener\">SIGIR18a<\/a>). We have also theoretically study weak supervision for information retrieval (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1321\" target=\"\\_blank\" rel=\"noopener\">ICTIR18<\/a>). Check out our SIGIR 2018 workshop on learning from limited or noisy data (<a href=\"https:\/\/lnd4ir.github.io\/\" target=\"\\_blank\" rel=\"noopener\">LND4IR<\/a>).<\/p>\n<p><strong>Conversational Information Seeking<\/strong>: I have been working on conversational search and recommendation with a focus on\u00a0<strong>mixed-initiative interactions<\/strong>, such as asking clarifying questions. We have designed and implemented an extensible platform for conversational information seeking research, called\u00a0<em>Macaw<\/em>\u00a0(<a href=\"https:\/\/github.com\/microsoft\/macaw\" target=\"\\_blank\" rel=\"noopener\">see the GitHub repository<\/a>). We have developed models for generating and selecting clarifying questions (<a href=\"http:\/\/hamedz.ir\/assets\/pub\/zamani-www2020.pdf\" target=\"\\_blank\" rel=\"noopener\">WWW20<\/a>,\u00a0<a href=\"https:\/\/arxiv.org\/pdf\/2006.00166.pdf\" target=\"\\_blank\" rel=\"noopener\">SIGIR20a<\/a>), models for utilizing user response to clarifying questions (<a href=\"https:\/\/arxiv.org\/pdf\/2006.07548.pdf\" target=\"\\_blank\" rel=\"noopener\">SIGIR20b<\/a>), and offline evaluation methodologies for asking clarifying questions in conversational systems (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1339\" target=\"\\_blank\" rel=\"noopener\">SIGIR19a<\/a>). I have recently started new directions related to\u00a0<strong>multi-modal conversational information seeking<\/strong>\u00a0(<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1430\" target=\"\\_blank\" rel=\"noopener\">SIGIR21<\/a>). We will participate in the Alexa Prize 2021 (<a href=\"http:\/\/ciir.cs.umass.edu\/alexaprize\" target=\"\\_blank\" rel=\"noopener\">see this news article<\/a>). Related to conversational search, we have implemented a unified search framework for mobile devices (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1301\" target=\"\\_blank\" rel=\"noopener\">SIGIR18b<\/a>,\u00a0<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1323\" target=\"\\_blank\" rel=\"noopener\">CIKM18b<\/a>).<\/p>\n<p><strong>IR Meets RecSys<\/strong>: I have made efforts on bridging the gap between the IR and RecSys communities. We have modeled query expansion as collaborative filtering (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1224\" target=\"\\_blank\" rel=\"noopener\">CIKM16<\/a>), proposed content-based recommendation models using IR techniques (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1282\" target=\"\\_blank\" rel=\"noopener\">RecSys17<\/a>,\u00a0<a href=\"https:\/\/link.springer.com\/article\/10.1007\/s10791-018-9327-0\" target=\"\\_blank\" rel=\"noopener\">IRJ18<\/a>), and designed a joint search and recommendation model (<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1320\" target=\"\\_blank\" rel=\"noopener\">DESIRES18<\/a>,\u00a0<a href=\"https:\/\/ciir-publications.cs.umass.edu\/pub\/web\/getpdf.php?id=1338\" target=\"\\_blank\" rel=\"noopener\">WSDM20<\/a>). I co-organized the ACM Recommender Systems Challenge in 2018 on music playlist generation and continuation, which can be seen as a RecSys or IR task (<a href=\"https:\/\/dl.acm.org\/citation.cfm?id=3240342\" target=\"\\_blank\" rel=\"noopener\">RecSys18<\/a>). Check out our visions on music recommendation (<a href=\"https:\/\/arxiv.org\/pdf\/1710.03208.pdf\" target=\"\\_blank\" rel=\"noopener\">IJMIR18<\/a>) and the overview of ACM RecSys Challenge 2018 (<a href=\"https:\/\/arxiv.org\/pdf\/1810.01520.pdf\" target=\"\\_blank\" rel=\"noopener\">TIST19<\/a>).<\/p>\n","protected":false},"excerpt":{"rendered":"<p>My research interests include various aspects of information retrieval, recommender systems, and text mining. My major contributions and current research focus include: Neural Information Retrieval: Following our initial study on distributed representations for information retrieval (ICTIR16a,\u00a0ICTIR16b,\u00a0CIKM16), we have developed multiple neural network models for the fundamental IR tasks. They include relevance-based word embedding (SIGIR17a), context-aware &hellip; <a href=\"https:\/\/groups.cs.umass.edu\/zamani\/research-topics\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Research Topics&#8221;<\/span><\/a><\/p>\n","protected":false},"author":67,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-196","page","type-page","status-publish","hentry","hfeed"],"_links":{"self":[{"href":"https:\/\/groups.cs.umass.edu\/zamani\/wp-json\/wp\/v2\/pages\/196","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/groups.cs.umass.edu\/zamani\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/groups.cs.umass.edu\/zamani\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/groups.cs.umass.edu\/zamani\/wp-json\/wp\/v2\/users\/67"}],"replies":[{"embeddable":true,"href":"https:\/\/groups.cs.umass.edu\/zamani\/wp-json\/wp\/v2\/comments?post=196"}],"version-history":[{"count":1,"href":"https:\/\/groups.cs.umass.edu\/zamani\/wp-json\/wp\/v2\/pages\/196\/revisions"}],"predecessor-version":[{"id":197,"href":"https:\/\/groups.cs.umass.edu\/zamani\/wp-json\/wp\/v2\/pages\/196\/revisions\/197"}],"wp:attachment":[{"href":"https:\/\/groups.cs.umass.edu\/zamani\/wp-json\/wp\/v2\/media?parent=196"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}