Yahoo! Labs

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Yahoo Labs[1] is the scientific engine guiding Yahoo innovation while powering impactful products for Yahoo’s users, partners, and advertisers.

Yahoo Labs includes approximately 200 research scientists and engineers, and serves as Yahoo’s research arm–its incubator for bold new ideas and laboratory for rigorous experimentation. Yahoo Labs applies its scientific findings in powering products for Yahoo’s users and enhancing value for its partners and advertisers. The Labs' forward-looking innovation also helps position Yahoo as an industry and scientific thought leader.

The mission statement of Yahoo Labs is "To power Yahoo's most critical products with innovative science."

Yahoo Labs:

  • Covers the spectrum from use-inspired foundational research to applied science
  • Anticipates and invent technology-based opportunities
  • Creates technology that is used by hundreds of millions of customers every day
  • Provides scientific insights on Yahoo's key growth areas: mobile, video, native advertising, and social
  • Explores fundamental computational, social, and economic phenomena
  • Works extremely closely with Yahoo software development organizations
  • Includes leaders in the international scientific research community
  • Works closely with the world’s leading universities

Yahoo Labs is headquartered in Sunnyvale, CA; it has three additional locations worldwide: New York, New York; London, England; Haifa, Israel.

History of Yahoo Labs[edit]

In the summer of 2005, a small group of leaders from some of the world’s most prominent industrial research institutions were afforded the privilege of starting something new at Yahoo–senior management had decided that the company was mature enough and the competitive landscape was challenging enough that Yahoo’s products needed to be more informed by, if not driven by, advanced scientific thinking and cutting-edge technology. And Yahoo Labs was born.

Yahoo Labs started out as Yahoo Research. In July of 2005, recruited by Usama Fayyad (formerly of Microsoft Research), who was Yahoo’s Chief Data Officer, Prabhakar Raghavan and Andrew Tomkins (both formerly of IBM Research) joined a few Yahoo employees (Kevin Lang among them) who were with Yahoo working for the erstwhile “Yahoo Research Labs” organization, and formed the brand new research team. Prabhakar was named as Head of Yahoo Research. Ronald J. Brachman, just finishing a term as office director at DARPA and a long career at Bell Labs and AT&T Labs, joined as the head of worldwide research operations in early September. The pioneering team was soon joined by a key set of additional leaders, including Andrei Broder and Ricardo Baeza-Yates. With that outstanding foundation, the group was able to build an incredible talent base of top scientists and research engineers in a remarkably short time.

In 2008 Yahoo Research merged with the Search and Advertising Applied Science organization and came to be known by its modern name, Yahoo Labs. In 2012, Ron Brachman took over as Head of Yahoo Labs and also became Yahoo’s chief scientist.

Over the years, the Labs became more and more essential to Yahoo’s products while maintaining a core scientific identity. The balance of the closeness of the team’s partnerships with Yahoo’s product and engineering groups, and the excellence of the external academic work achieved, is unparalleled in the industry. The research scientists and engineers at Yahoo Labs contribute critical insights to Yahoo’s most important products; in many cases teams are actually fully embedded in the workspaces of their product partners. All the while the Labs continues to maintain extraordinary standards of technical excellence through rigorous interviews with candidates, top-notch publications in the best venues, and senior leadership positions in various scientific fields.

Research areas[edit]

Yahoo Labs research scientists and engineers work across twelve broad research areas:

Advertising Science[edit]

Advertising Science experts at Yahoo Labs work closely with Yahoo's advertising product and engineering teams to monetize Yahoo's and publishers’ content by connecting advertisers and users in the right context. We use large-scale statistical modeling, data mining, and optimization techniques to develop algorithms for serving the most relevant ads to users and provide value to advertisers. We also understand marketplaces and mechanism design, auctions, and other important related economic phenomena, as well as understanding users and advertisers through data analytics.

Computer Vision[edit]

Computer Vision research at Yahoo Labs focuses on all aspects of image and video content understanding. We work on a variety of problems ranging from automatic classification and labeling of content to image and video aesthetics. Our research is human-centered, which means that the work is carried out at the intersection of large-scale data analysis, interaction, and human aspects in the creation and consumption of multimedia content, with a strong focus on user behavior. We apply machine learning to build tools and create algorithms that impact many of Yahoo's flagship products. Since we place a strong emphasis on users, our work includes research in social media and recommender systems.

Human-Computer Interaction[edit]

Human-Computer Interaction research at Yahoo Labs aids in building the next generation of Internet experiences by investigating the quantitative and qualitative aspects of how people communicate using social networking sites and multimedia. Primarily partnered with Flickr, we use a set of mixed-methods, including social and visual analytics, information visualization, and interview studies to investigate online photography, blogs, personal collections, and community systems. We provide expertise in the areas of user-generated content, data analysis, search, recommendation systems, and geographic computing.

Information and Knowledge Management[edit]

Information and Knowledge Management research at Yahoo Labs includes a strong engineering component to maintain and enrich the Yahoo Knowledge graph, a large unified knowledge structure that powers a wide variety of knowledge-based services at Yahoo. Information and knowledge management research focuses on creating innovative solutions for turning siloed, incomplete, inconsistent, and potentially inaccurate information extracted from heterogenous data sources into a rich, unified, disambiguated knowledge base consisting of millions of interconnected entities. Our research touches multiple fields, including Machine Learning, Data Mining, and Knowledge Representation, all at unprecedented scale.

Machine Learning and Data Science[edit]

Machine Learning and Data Science efforts at Yahoo Labs focus on research and development of production-quality solutions to important data-intensive problems at Yahoo, including algorithms, frameworks and tools for large scale machine learning, data visualization, metrics reporting, and analytics. We live and breathe Big Data every day, and work to apply the most leading-edge machine learning technologies to Yahoo's toughest large-scale problems.

Mail Search and Analysis[edit]

Mail Search and Analysis is critical for Yahoo, as Yahoo Mail represents the majority of Yahoo's pageviews. Furthermore, inboxes have turned into repositories of valuable personal information, from travel itineraries and bills to family photos. Consequently mail search is evolving to include personal content results across multiple platforms. While relying on similar foundations to Web search, Mail Search scientists at Yahoo Labs face additional challenges, including privacy-preserving data processing and more drastic demand for perfect quality than Web search, as users know their own content. At the same time there are new research opportunities, including mining mail data at scale with its majority of machine-generated messages mixed with personal communication. This is a new frontier of research that can only be explored at this scale by a few companies like Yahoo.

Metrics and User Engagement[edit]

Metrics and User Engagement research at Yahoo Labs is closely aligned with Search, Personalization, Knowledge, and Ads to study and define metrics for experimentation, user engagement, and product success. Statistical analysis and data-driven user understanding methods are employed to enable the optimization of products through rigorous experimentation. Researchers help guide Yahoo product development by providing deep, grounded engagement metrics to ensure a focus on a user-centric approach to long-term product health. Metrics and user engagement research also aids Yahoo as a whole in monitoring data integrity and clean instrumentation.

Natural Language Processing and Information Extraction[edit]

Natural Language Processing and Information Extraction in Yahoo Labs involves close collaboration with the core content processing platforms across Search, Personalization, Knowledge, and Ads. The focus is on creating innovative, Web-scale solutions for the whole range of content processing including content representation, classification and clustering, text mining, structured and unstructured information extraction, and summarization. A key aim of the research is to enable fully- or semi-automated methods that both scale to the Web but also self-correct and improve automatically over time.

Personalization and Contextual Search[edit]

At Yahoo Labs, Personalization and Contextual Search joins traditional search and recommendation with the emerging idea of personalization within real-time experiences on mobile devices using rich context, adding anticipation to the mix. Anticipation depends on ""hyper-personalized"" and ""hyper-contextualized"" data, for example, as in predicting in real time that a person is about to begin their nightly jogging routine. A system could then open the user's favorite jogging app and execute the “start a run” action, and at the same time open a favorite music app and tune to a well-suited local radio station. Key problems in this area include contextual profiling, context-based intent prediction, and real-time federated search.

Security and Privacy[edit]

Security and Privacy research at Yahoo Labs partners with Yahoo's Membership, Security Engineering, and Threat Response teams to protect Yahoo and its users from hackers, spammers, and other hostile Web entities. We address and fix urgent security needs, design and help deploy sound security solutions, and advance the state-of-the-art in security and privacy. Our research covers a broad range of topics in information security, including design and implementation of advanced authentication schemes, cryptographic primitives and protocols and their applications, and statistical data privacy and its relation to machine learning and data mining.

Systems[edit]

Systems research at Yahoo Labs enables high performance, efficient, reliable, scalable cloud and mobile platforms through deep understanding of Yahoo infrastructure and product needs. Primarily partnered with Yahoo's Platforms & Personalization Products organization, Systems researchers help guide the architectures of the internal and open-source, software and hardware systems at Yahoo. We also build the systems, techniques, and tools necessary for scalable performance of current and future products.

Web Search and Data Mining[edit]

Web Search and Data Mining research at Yahoo Labs is focused on the four key areas of content acquisition, content understanding, query analysis, and content ranking. As part of content acquisition, our research focuses on efficient means for crawling the Web, ensuring freshness, and performing topic-focused crawls or deep crawls. Content understanding focuses on document structural analysis, information extraction, categorization and clustering, etc. Query analysis and content ranking focus on how to predict user intent and how to match and rank the most relevant information satisfying user needs.

Groups unrelated to research[edit]

Yahoo Labs includes two groups unrelated to research:

Academic Partnerships[edit]

Academic Partnerships at Yahoo Labs maximizes the impact of Yahoo’s engagements with the global scientific community by building deep relationships with select academic research institutions and fostering collaborative research through dynamic programs: sharing large-scale datasets for research, managing Yahoo's donation of computing resources, supporting diversity and campus events, hosting academic visits and talks, and funding project grants and merit-based awards. The Academic Partnerships team positions Yahoo as one of the first companies that leading academics think of for new research collaborations, employment opportunities, and inspiring and entertaining online products. Additionally, the team aids Yahoo with graduate student sourcing and helps keep the company up-to-date on strategic opportunities and insights from the external research community.

Operations[edit]

Operations at Yahoo Labs is responsible for managing the day-to-day processes and procedures for all of the organization's research scientists and engineers across the world. The Operations team leads Yahoo corporate initiatives on behalf of the Labs and tracks progress for Labs goals. Furthermore, the Operations team is responsible for managing Yahoo Labs' finances, recruiting efforts, events, and communications. Perhaps most importantly, the team nurtures, grows, and supports Yahoo Labs' culture in order to retain and attract the top researchers in industry and academia.

Notable accomplishments[edit]

Social Media[edit]

Updates from Yahoo Labs can be followed on the Yahoo Labs website, https://labs.yahoo.com, and on the following social media channels:

References[edit]

External links[edit]