Mendeley: Recommendations for Researchers
by Saúl Vargas (Mendeley)
For a researcher, keeping up with what is going on in their research field can be a difficult and time-consuming task. Since the volume of published research and research activity is constantly growing, it is becoming increasingly more difficult for researchers to be able to manage and filter through the research information flow. In this challenging context, Mendeley’s mission is to become the world’s “research operating system”. We do this not only by providing our well-know reference management system, but also by providing discovery capabilities for researchers on different kinds of entities, such as articles and profiles. In our talk, we will share Mendeley’s experiences with building our article and profile recommendation systems, the challenges that we have faced and the solutions that we have put in place. We will discuss how we address different users’ needs with our data and algorithm infrastructure to achieve good user experience.
About the Speaker
Saúl Vargas is a Senior Data Scientist at Mendeley and Elsevier, where he is working on a variety of tools and products powering recommendations for researchers. Before that, he was working as a postdoc at the University of Glasgow doing research in the area of Social Media Analysis. He has a PhD in the area of recommender systems from the Autonomous University of Madrid under the supervision of Prof. Pablo Castells.
When Recommendation Systems Go Bad
by Evan Estola (Meetup)
Machine learning and recommendations systems have changed the way we interact with not just the internet, but some of the basic services that we use to organize and run our life. As the people that build these systems, we have a social responsibility to consider how these systems affect people, and furthermore, we should do whatever we can to prevent these models from perpetuating some of the prejudice and bias that exist in our society today. This talk will cover some of the recommendation systems that have gone wrong across various industries, and attempt to provide some solutions for raising awareness and prevention. Approaches that will be explored include using interpretable models, using ensemble models to separate features that shouldn’t interact, and designing test data sets for capturing accidental bias.
About the Speaker
Evan is a Lead Machine Learning Engineer working on the Data Team at Meetup. Combining product design, machine learning research and software engineering, Evan builds systems that help Meetup’s members find the best thing in the world: real local community. Before Meetup, Evan worked on hotel recommendations at Orbitz Worldwide, and he began his career in the Information Retrieval Lab at the Illinois Institute of Technology.
News Recommendations at Scale at Bloomberg Media: Challenges and Approaches
by Dhaval Shah & Rohit Parimi (Bloomberg)
In the past decade, news consumption via traditional channels such as print has been on the decline while online and digital news consumption has been steadily growing. Bloomberg, renowned for its products in the financial world, has a very strong presence in the news and media industry. Bloomberg Media, on an average, publishes 400-500 stories and videos per day and we have close to 30 million unique visitors on our websites and mobile applications every month consuming this content. At such a scale it is very important to recommend relevant information for a good user experience. Recommendations in the News and Media domain bring a unique set of challenges due to the dynamic nature of the data as well as unique consumption patterns.
The biggest challenge with building recommendation systems in the News domain is the dynamic nature of the domain itself; new content is published every few minutes and majority of the content has a short shelf life, i.e., the news is not relevant to users after a certain time span and the time span is generally of the order of hours rather than days, making it important to deliver relevant content in a timely manner. Moreover, our users consume content differently based on time of day. For example, some users whose focus is market news and market data during the day consume more long form and generic articles and videos in the evening. User preferences, along with being cyclical in nature, tend to change over time, so algorithms need to adapt to the changing taste of the user. In addition, we need to ensure that the users do get their share of important/trending news and are not put into a filter bubble.
In this talk, we will present some novel techniques we have applied to popular approaches in the field of Recommender Systems to be able to address the unique challenges which the news domain presents.
About the Speakers
Dhaval Shah is an Engineering Manager at Bloomberg, overseeing all Machine Learning and Big Data related activities for Bloomberg Media. Dhaval’s team has been applying Machine Learning and Big Data techniques to build Recommender Systems for Bloomberg Media’s websites and native applications, including the flagship Bloomberg.com website, since 2011. Dhaval’s team is also responsible for building automated tagging systems for Bloomberg’s News articles as well as systems for personalizing push notifications for Bloomberg’s iOS and Android applications.
Rohit Parimi is a Software Engineer at Bloomberg, since 2015. He is part of the Machine Learning and Big Data team at Bloomberg Media and works primarily on recommender and tagging systems for news content. He has a PhD from Kansas State University, focused on cross-domain recommender systems.
Marsbot: Building a Personal Assistant
by Max Sklar (Foursquare)
Foursquare recently launched Marsbot, an SMS-based app for local recommendations. Marsbot is an intelligent friend that lives in your pocket and learns about you through the places you go in the real world. While this product is aligned with Foursquare’s long-standing mission to find the best places, it represents a new era in the way people interact with recommendation engines. The promise of the latest crop of personal assistants is get us information more quickly and seamlessly, but building them comes with many challenges. In this talk, we discuss why we built Marsbot and some of the many lessons learned along the way.
About the Speaker
Max Sklar is a data scientist and engineer at Foursquare and frequent contributor to RecSys. As part of the engineering team, Max has focused on using both machine learning and heuristics to build new features into the apps Foursquare and Swarm. His contributions include work on early versions of the recommendation algorithm, venue ratings, and the natural language processing stack. Most recently, Max pulled out all stops in the organization to launch of Marsbot, the topic of his talk. He has presented at a variety of conferences, including the 2015 Workshop on Urban Data Science at Cambridge University, and is a frequent guest speaker at meetups and university courses. He holds an M.S. in Information Systems from NYU, and a B.S. in Computer Science from Yale.
Music Personalization at Spotify
by Vidhya Murali (Spotify)
Spotify is the world’s largest on-demand music streaming company, with over 75 million active users choosing what to listen to among tens of millions of songs. Discovery and personalization play a key role in enhancing the users’ experience on Spotify. In this talk, we’ll discuss how Spotify uses collaborative filtering and content signals to model users’ tastes. We will dive deep into the workings of features such as Discover Weekly, Fresh Finds and Home that are powered by our machine learning models. We will also discuss the challenges around optimizing for user engagement and scaling for our growing user base and music catalog.
About the Speaker
Vidhya is an engineer at Spotify’s Music Discovery and Personalization team. She has domain interests in big data and machine learning and has a passion for building products. She works on building scalable data pipelines and machine learning models to power users discover their next favorite music. She received her Masters in Computer Science from the University of Wisconsin-Madison and Bachelors from BITS Pilani, India.
