{"id":489929,"date":"2018-06-11T09:08:49","date_gmt":"2018-06-11T16:08:49","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-event&p=489929"},"modified":"2018-07-23T21:08:56","modified_gmt":"2018-07-24T04:08:56","slug":"microsoft-sigir-2018","status":"publish","type":"msr-event","link":"https:\/\/www.microsoft.com\/en-us\/research\/event\/microsoft-sigir-2018\/","title":{"rendered":"Microsoft @ SIGIR 2018"},"content":{"rendered":"
Venue:<\/strong> Michigan League (opens in new tab)<\/span><\/a> (Location on Campus Map (opens in new tab)<\/span><\/a>)<\/p>\n Website:<\/strong> SIGIR 2018 (opens in new tab)<\/span><\/a><\/p>\n","protected":false},"excerpt":{"rendered":" Venue: Michigan League (Location on Campus Map) Website: SIGIR 2018<\/p>\n","protected":false},"featured_media":490106,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"msr_startdate":"2018-07-08","msr_enddate":"2018-07-12","msr_location":"Ann Arbor, Michigan, USA","msr_expirationdate":"","msr_event_recording_link":"","msr_event_link":"http:\/\/sigir.org\/sigir2018\/attend\/","msr_event_link_redirect":false,"msr_event_time":"","msr_hide_region":true,"msr_private_event":false,"footnotes":""},"research-area":[13556,13555],"msr-region":[197900],"msr-event-type":[197941],"msr-video-type":[],"msr-locale":[268875],"msr-program-audience":[],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-489929","msr-event","type-msr-event","status-publish","has-post-thumbnail","hentry","msr-research-area-artificial-intelligence","msr-research-area-search-information-retrieval","msr-region-north-america","msr-event-type-conferences","msr-locale-en_us"],"msr_about":"Venue:<\/strong> Michigan League<\/a> (Location on Campus Map<\/a>)\r\n\r\nWebsite:<\/strong> SIGIR 2018<\/a>","tab-content":[{"id":0,"name":"About","content":"SIGIR is a major international forum for presentation of the latest state-of-the-art research and demonstration of new systems and methods for connecting people with information: from Web search engines, recommender systems, and social network technology to compelling applications in health, legal, educational, and other domains, research at SIGIR spans both academia and industry.\r\n AI & Research (AI&R) at Hyderabad, India comprises of highly motivated researchers, engineers, product managers and data-scientists building end-to-end web-scale and enterprise-scale AI systems. We seek talented, energetic, creative and passionate ML engineers with ability to enhance and apply research to ship and build high-quality products and services.<\/p>"},{"id":1,"name":"Accepted Papers","content":"Program Committee members<\/h2>\r\nPaul Bennett<\/a>, Short Paper Chair\r\nJianfeng Gao<\/a>, AI Track Co-chair\r\n
Invited Speakers<\/h2>\r\nDistributional Representation of Complex Semantics<\/strong> (Keynote at KG4IR workshop<\/a>)\r\nKuansan Wang<\/a>, Microsoft Research\r\n\r\nLessons from Building a Large-scale Commercial IR-based Chatbot for an Emerging Market<\/strong>\r\nPuneet Agrawal and Manoj Kumar Chinnakotla, Microsoft\r\n\r\nCausal Inference over Longitudinal Data to Support Expectation Exploration<\/strong>\r\nEmre Kiciman<\/a>, Microsoft Research\r\n\r\nSearch and Recommendation in the Enterprise<\/strong><\/a>\r\nPaul Bennett<\/a>, Microsoft Research\r\n
Workshops<\/h2>\r\nLearning from Limit\/Noisy data for IR<\/a>\r\nHamed Zamani (UMass Amherst), Mostafa Dehghani (Univ. of Amsterdam), Fernando Diaz (Microsoft Research \u2013 Montreal), Hang Li (Toutiao AI Lab), Nick Craswell (Microsoft)\r\n
Microsoft attendees<\/h2>\r\nAmjad Abu-Jbara, Microsoft\r\nOmar Alonso, Microsoft\r\nAhmed Awadallah<\/a>, Microsoft Research\r\nPaul Bennett<\/a>, Microsoft Research\r\nEdward Cui, Microsoft\r\nWeiwei Deng, Microsoft\r\nFernando Diaz, Microsoft Research \u2013 Montreal\r\nSusan Dumais<\/a>, Microsoft Research\r\nAdam Fourney<\/a>, Microsoft Research AI\r\nEmre Kiciman<\/a>, Microsoft Research\r\nXiaoliang Ling, Microsoft\r\nPawel Pietrusinski, Microsoft\r\nMona Soliman Habib, Microsoft\r\nHui Su, Microsoft\r\n
Career Opportunities<\/h2>\r\n
ML Engineer<\/a><\/h4>\r\n
Full Papers<\/h2>\r\nCalendar-Aware Proactive Email Recommendation<\/strong>\r\nQian Zhao (University of Minnesota); Paul Bennett (Microsoft); Adam Fourney (Microsoft); Anne Thompson (Microsoft); Shane Williams (Microsoft); Adam D. Troy (Microsoft); Susan Dumais (Microsoft)\r\n\r\nCharacterizing and Supporting Question Answering in Human-to-Human Communication<\/strong>\r\nXiao Yang (The Pennsylvania State University); Ahmed Hassan Awadallah (Microsoft); Madian Khabsa (Apple); Wei Wang (Microsoft); Miaosen Wang (Microsoft)\r\n\r\nDeep Domain Adaptation Hashing with Adversarial Learning<\/strong>\r\nFuchen Long (University of Science and Technology of China); Ting Yao (Microsoft); Qi Dai (Microsoft); Xinmei Tian (University of Science and Technology of China); Jiebo Luo (University of Rochester); Tao Mei (Microsoft)\r\n\r\nMeasuring the Utility of Search Engine Result Pages<\/strong>\r\nLeif Azzopardi (University of Strathclyde); Paul Thomas (Microsoft); Nick Craswell (Microsoft)\r\n\r\nNatural Language Interfaces with Fine-Grained User Interaction: A Case Study on Web APIs<\/strong>\r\nYu Su (University of California Santa Barbara); Ahmed Hassan Awadallah (Microsoft); Miaosen Wang (Microsoft); Ryen White (Microsoft)\r\n\r\nTowards Better Text Understanding and Retrieval through Kernel Entity Salience Modeling<\/strong>\r\nChenyan Xiong (Carnegie Mellon University); Zhengzhong Liu (Carnegie Mellon University); Jamie Callan (Carnegie Mellon University); Tie-Yan Liu (Microsoft)\r\n
Short Papers<\/h2>\r\nAd Click Prediction in sequence with Long Short-Term Memory Networks: An externality-aware model<\/strong>\r\nWeiwei Deng (Microsoft); Xiaoliang Ling (Microsoft); Yang Qi (Microsoft); Tunzi Tan (School of Mathematical Sciences @ University of Chinese Academy of Sciences); Eren Manavoglu (Microsoft); Qi Zhang (Microsoft)\r\n\r\nAssessing the Readability of Web Search Results for Searchers with Dyslexia<\/strong>\r\nAdam Fourney (Microsoft); Meredith Ringel Morris (Microsoft); Abdullah Ali (University of Washington); Laura Vonessen (University of Washington)\r\n\r\nAttention-driven Factor Model for Explainable Personalized Recommendation<\/strong>\r\nJingwu Chen (Institute of Computing Technology, Chinese Academy of Sciences); Fuzhen Zhuang (Institute of Computing Technology, Chinese Academy of Sciences); Xin Hong (Institute of Computing Technology, Chinese Academy of Sciences); Xiang Ao (Institute of Computing Technology, Chinese Academy of Sciences); Xing Xie (Microsoft); Qing He (Institute of Computing Technology, Chinese Academy of Sciences)\r\n\r\nCross Domain Regularization for Neural Ranking Models using Adversarial Learning<\/strong>\r\nDaniel Cohen (University of Massachusetts Amherst); Bhaskar Mitra (Microsoft); Katja Hofmann (Microsoft); Bruce Croft (University of Massachusetts Amherst)\r\n\r\nMulti-level Abstraction Convolutional Model with Weak Supervision for Information Retrieval<\/strong>\r\nYifan Nie (University of Montreal); Alessandro Sordoni (Maluuba \u2013 Microsoft); Jian-Yun Nie (University of Montreal)\r\n\r\nOptimizing Query Evaluations using Reinforcement Learning for Web Search<\/strong>\r\nCorby Rosset (Microsoft); Damien Jose (Microsoft); Gargi Ghosh (Microsoft); Bhaskar Mitra (Microsoft); Saurabh Tiwary (Microsoft)\r\n\r\nQuantitative Information Extraction From Social Data<\/strong>\r\nOmar Alonso (Microsoft); Thibault Sellam (Columbia University)\r\n\r\nTesting the Cluster Hypothesis with Focused and Graded Relevance Judgments<\/strong>\r\nEilon Sheetrit (Technion \u2013 Israel Institute of Technology); Anna Shtok (Technion \u2013 Israel Institute of Technology); Oren Kurland (Technion, Israel Institute of Technology); Igal Shprincis (Microsoft, Herzliya, Israel)\r\n\r\nTransparent Tree Ensembles<\/strong>\r\nAlexander Moore (Microsoft); Vanessa Murdock (Microsoft); Yaxiong Cai (Microsoft); Kristine Jones (Microsoft)\r\n
SIRIP Industry Papers<\/h2>\r\nPuneet Agrawal and Manoj Kumar Chinnakotla. Lessons from Building a Large-scale Commercial IR-based Chatbot for an Emerging Market<\/strong>\r\nPuneet Agrawal (Microsoft); Manoj Kumar Chinnakotla (Microsoft)"},{"id":2,"name":"Conference Analytics","content":"The Microsoft Academic Graph<\/a> makes it possible to gain analytic insights about any of the entities within it: publications, authors<\/a>, institutions<\/a>, topics<\/a>, journals<\/a>, and conferences<\/a>. Below, we present historical trend analysis about the SIGIR\u2013 Special Interest Group on Information Retrieval\u2013Conference.\r\n\r\nYou can generate your own insights by accessing the Microsoft Academic Graph through the Academic Knowledge API<\/a> or through Azure Data Lake Store<\/a> (please contact us<\/a> for the latter option). If you would like to learn how we generated the insights below, please see the repository with source code<\/a>.\r\n\r\nClick on each image for current trends and data hosted by Microsoft Academic Graph<\/a>.<\/em>\r\n
SIGIR paper output<\/h2>\r\nThe chart below shows the evolution of the number of conference papers for each conference year.\r\n\r\n<\/a>\r\n\r\nIn the following chart, the black bars represent average numbers of references per conference paper for each year. The data show that recent publications tend to cite more references. The green bars show the average number of citations of conference papers written in a given year. Note that the citations are raw counts and not normalized by the age of publications. This is because the \u201ccorrect\u201d way to normalize the citation counts turns out to be a nontrivial problem and may well be application dependent. Please treat the raw data presented as an invitation to conduct research on this topic!\r\n\r\n<\/a>\r\n\r\nThat being said, a visible trend is that older publications tend to receive more citations because they have more time for researchers to recognize the contributions of the paper. There are, however, notable exceptions, the first in 1994, due to several highly cited papers:\r\n
\r\n \t
Memory of references<\/h2>\r\nHow old are the papers cited by SIGIR papers? Follow a given year\u2019s column to see the age of papers cited in conference papers published that year. For example, in 2017, SIGIR papers collectively cited 683 papers published in 2016, 657 papers published in 2015, and so on.\r\n\r\n<\/a>\r\n\r\n*If some years appear to cite publications from the future, it is most likely because they cited books. When a new edition of the book appeared, it replaced the previous one in the Microsoft Academic Graph and the citation appears to be from the future. In this representation, to generate a cleaner view, we removed all instances of papers citing papers more than two years in the future.<\/em>\r\n
Outgoing references<\/h2>\r\nWhat venues do SIGIR papers cite?\r\n\r\nThe pie chart shows the top 10 venues cited by\u00a0SIGIR papers over time.\u00a0SIGIR, CIKM, and WWW emerge as the top three.\r\n\r\n<\/a>\r\n\r\nThe 100 percent stacked bar chart below shows the percent of references given by SIGIR\u00a0papers to each of the top 20 venues, year by year.\r\n\r\n<\/a>\r\n
Incoming citations<\/h2>\r\nWhat venues cite SIGIR papers?\r\n\r\nThe pie chart below shows the top 10 venues of all time that cite SIGIR papers.\u00a0 SIGIR is the top one, followed by CIKM, and Information Processing and Management. See the table for year-by-year details of citations coming from each of the top 10 venues.\r\n\r\n<\/a>\r\n\r\nThe 100 percent stacked bar chart below shows the citation distribution from the top 20 citing venues, year by year.\r\n\r\n<\/a>\r\n
Most-cited authors<\/h2>\r\nWho are the most-cited authors of all time in SIGIR papers? The interactive chart below ranks the most-cited authors by using number of publications cited by the conference and number of citations received from the conference. Authors do not have to have published in SIGIR to appear on this chart.\r\n\r\n<\/a>\r\n\r\nWho are the rising stars among the top cited authors in SIGIR? The 100 percent stacked bar chart below shows the citation distribution by the top 20 authors, year by year.\r\n\r\n<\/a>\r\n
Top institutions<\/h2>\r\nThe bubble chart visualizes the top institutions at SIGIR by citation count. The size of the bubble is proportional to the total number of publications from that institution at SIGIR.\r\n\r\n<\/a>\r\n\r\nGet the most current data and also explore the top institutions at the conference in more detail by clicking the chart. Once on the underlying Microsoft PowerBI dashboard, click on a column to rank the top institutions by publication or citation count.\r\n\r\n<\/a>\r\n
Top authors<\/h2>\r\nThe next three charts show author rankings according to different criteria.\r\n\r\nThe bubble chart displays SIGIR authors ranked by citation count, with bubble size being relative to publication count.\r\n\r\n<\/a>\r\n\r\nGet the most current data and also explore the top authors at the conference in more detail by clicking the chart. Once on the underlying Microsoft PowerBI dashboard, you can also explore the top conference authors in more detail. Click on a column to rank the top authors by Microsoft Academic rank, publication, or citation count.\r\n\r\n<\/a>\r\n\r\nThe bubble chart below visualizes author rank, which is calculated by Microsoft Academic by using a formula that is less susceptible to citation counts than similar measures. The X axis shows author rank. The higher an author\u2019s rank, the closer they are to the right side. The Y axis normalizes the rank by publication count and enables us to identify impactful authors who might not have had a very large number of publications. The closer an author is to the top, the higher their normalized rank. Of course, the area of the chart that represents the highest rank is the top right corner.\r\n\r\n<\/a>\r\n\r\nStephen Roberson is an interesting case. Although he is one of the most influential authors in the information retrieval field, he\u2019s only ranked at the 19th<\/sup> place for SIGIR conference. It turns out the Stephen\u2019s best work is not published at SIGIR. BM25F is published at CIKM in 2004 [1], then in a booklet in 2009 [2]. He got his fame mostly from Okapi, published first at 1994 TREC [3] through 1999 [4], again, at TREC. His most well-cited work at SIGIR is an approximation to 2-Poisson model [5], and a CAL paper with the Bing team using pseudo-relevance feedback [6] that is no longer in the production. He co-authored a paper questioning the use of language modeling techniques for IR [7] which, unfortunately, prevailed until today against his predictions.\r\n
\r\n \t