Archive for the ‘Microsoft Research’ Category

“They’re blowing up my phone”: Group Messaging Practices Among Adolescents

July 13, 2015 Comments off

“They’re blowing up my phone”: Group Messaging Practices Among Adolescents
Source: Microsoft Research

While group messaging has become popular, particularly among adolescents, it has not yet been explored in the HCI literature. We interviewed 48 adolescents, aged 15-24, who use group messaging regularly. We present a framework for understanding the types of groups they communicate with according to three dimensions: focus, membership, and duration. We also present findings about factors influencing their choice of group messaging tools and the problems they have managing multiple group threads using multiple tools. We explore the problem of notification overload and users’ strategies for managing frequent notifications. We describe one approach of “social alerting, ” when group members notify one another directly, rather than rely on app notifications. We relate our findings to prior work and offer design suggestions to address the challenges our participants faced in managing frequent group notifications.

An Overview of Microsoft Academic Service (MAS) and Applications

July 10, 2015 Comments off

An Overview of Microsoft Academic Service (MAS) and Applications
Source: Microsoft Research

In this paper we describe a new release of a Web scale entity graph that serves as the backbone of Microsoft Academic Service (MAS), a major production effort with a broadened scope to the namesake vertical search engine that has been publicly available since 2008 as a research prototype. At the core of MAS is a heterogeneous entity graph comprised of six types of entities that model the scholarly activities: field of study, author, institution, paper, venue, and event. In addition to obtaining these entities from the publisher feeds as in the previous effort, we in this version include data mining results from the Web index and an in-house knowledge base from Bing, a major commercial search engine. As a result of the Bing integration, the new MAS graph sees significant increase in size, with fresh information streaming in automatically following their discoveries by the search engine. In addition, the rich entity relations included in the knowledge base provide additional signals to disambiguate and enrich the entities within and beyond the academic domain. The number of papers indexed by MAS, for instance, has grown from low tens of millions to 83 million while maintaining an above 95% accuracy based on test data sets derived from academic activities at Microsoft Research. Based on the data set, we demonstrate two scenarios in this work: a knowledge driven, highly interactive dialog that seamlessly combines reactive search and proactive suggestion exper

Inverse Privacy

July 7, 2015 Comments off

Inverse Privacy
Source: Microsoft Research

An item of your personal information is inversely private if some party has access to it but you do not. We analyze the provenance of inversely private information and its rise to dominance over other kinds of personal information. In a nutshell, the inverse privacy problem is unjustified inaccessibility to you of your inversely private information. We believe that the inverse privacy problem has a market-based solution.

Large Scale Log Analysis of Individuals’ Domain Preferences in Web Search

June 12, 2015 Comments off

Large Scale Log Analysis of Individuals’ Domain Preferences in Web Search
Source: Microsoft Research

Information on almost any given topic can be found on the Web, often accessible via many different websites. But even when the topical content is similar across websites, the websites can have different characteristics that appeal to different people. As a result, individuals can develop preferred websites to visit for certain topics. While it has long been speculated that such preferences exist, little is understood about how prevalent, clear, and stable these preferences actually are. We characterize website preference in search by looking at repeat domain use in two months of large-scale query and webpage visitation logs. We show that while people sometimes provide explicit cues in their queries to indicate their domain preferences, there is a significant opportunity to identify implicit preferences expressed via user behavior. Although domain preferences vary across users, within a user they are consistent and stable over time, even during events that typically disrupt normal search behavior. People’s preferences do, however, vary given the topic of their search. We observe that people exhibit stronger domain preferences while searching than browsing, but that search-based preferences often extend to pages browsed to after the initial search result click. Since domain preferences are common for search and stable over time, the rich understanding of them that we present here will be valuable for personalizing search.

Categories: Microsoft Research, search

Questions vs. Queries in Informational Search Tasks

May 29, 2015 Comments off

Questions vs. Queries in Informational Search Tasks
Source: Microsoft Research

Search systems traditionally require searchers to formulate information needs as keywords rather than in a more natural form, such as questions. Recent studies have found that Web search engines are observing an increase in the fraction of queries phrased as natural language. As part of building better search engines, it is important to understand the nature and prevalence of these intentions, and the impact of this increase on search engine performance. In this work, we show that while 10.3% of queries issued to a search engine have direct question intent, only 3.2% of them are formulated as natural language questions. We investigate whether search engines perform better when search intent is stated as queries or questions, and we find that they perform equally well to both.

Categories: Microsoft Research, search

Information Retrieval with Verbose Queries

May 23, 2015 Comments off

Information Retrieval with Verbose Queries
Source: Microsoft Research

Recently, the focus of many novel search applications shifted from short keyword queries to verbose natural language queries. Examples include question answering systems and dialogue systems, voice search on mobile devices and entity search engines like Facebook’s Graph Search or Google’s Knowledge Graph. However the performance of textbook information retrieval techniques for such verbose queries is not as good as that for their shorter counterparts. Thus, effective handling of verbose queries has become a critical factor for adoption of information retrieval techniques in this new breed of search applications. Over the past decade, the information retrieval community has deeply explored the problem of transforming natural language verbose queries using operations like reduction, weighting, expansion, reformulation and segmentation into more effective structural representations. However, thus far, there was not a coherent and organized tutorial on this topic. In this tutorial, we aim to put together various research pieces of the puzzle, provide a comprehensive and structured overview of various proposed methods, and also list various application scenarios where effective verbose query processing can make a significant difference.

The Emerging Role of Data Scientists on Software Development Teams

April 20, 2015 Comments off

The Emerging Role of Data Scientists on Software Development Teams
Source: Microsoft Research

Creating and running software produces large amounts of raw data about the development process and the customer usage, which can be turned into actionable insight with the help of skilled data scientists. Unfortunately, data scientists with the analytical and software engineering skills to analyze these large data sets have been hard to come by; only recently have software companies started to develop competencies in software-oriented data analytics. To understand this emerging role, we interviewed data scientists across several product groups at Microsoft. In this paper, we describe their education and training background, their raison d’être in software engineering contexts, and the type of problems on which they work. We identify five distinct working styles of data scientists and describe a set of strategies that they employ to increase the impact and actionability of their work.