Martin Saveski

Item cold-start recommendations: learning local collective embeddings

Recommender systems suggest to users items that they might like (e.g., news articles, songs, movies) and, in doing so, they help users deal with information overload and enjoy a personalized experience. One of the main problems of these systems is the item cold-start, i.e., when a new item is introduced in the system and no past information is available, then no effective recommendations can be produced. The item cold-start is a very common problem in practice: modern online platforms have hundreds of new items published every day. To address this problem, we propose to learn Local Collective Embeddings: a matrix factorization that exploits items properties and past user preferences while enforcing the manifold structure exhibited by the collective embeddings. We present a learning algorithm based on multiplicative update rules that are efficient and easy to implement. The experimental results on two item cold-start use cases: news recommendation and email recipient recommendation, demonstrate the effectiveness of this approach and show that it significantly outperforms six state-of-the-art methods for item cold-start.

Detecting Network Effects: Randomizing Over Randomized Experiments

Randomized experiments, or A/B tests, are the standard approach for evaluating the causal effects of new product features, i.e., treatments. The validity of these tests rests on the stable unit treatment value assumption (SUTVA), which implies that the treatment only affects the behavior of treated users, and does not affect the behavior of their connections. Violations of SUTVA, common in features that exhibit network effects, result in inaccurate estimates of the causal effect of treatment. In this paper, we leverage a new experimental design for testing whether SUTVA holds, without making any assumptions on how treatment effects may spill over between the treatment and the control group. To achieve this, we simultaneously run both a completely randomized and a cluster-based randomized experiment, and then we compare the difference of the resulting estimates. We present a statistical test for measuring the significance of this difference and offer theoretical bounds on the Type I error rate. We provide practical guidelines for implementing our methodology on large-scale experimentation platforms. Importantly, the proposed methodology can be applied to settings in which a network is not necessarily observed but, if available, can be used in the analysis. Finally, we deploy this design to LinkedIns experimentation platform and apply it to two online experiments, highlighting the presence of network effects and bias in standard A/B testing approaches in a real-world setting.

One-Pass Ranking Models for Low-Latency Product Recommendations

Purchase logs collected in e-commerce platforms provide rich information about customer preferences. These logs can be leveraged to improve the quality of product recommendations by feeding them to machine-learned ranking models. However, a variety of deployment constraints limit the naive applicability of machine learning to this problem. First, the amount and the dimensionality of the data make in-memory learning simply not possible. Second, the drift of customers preference over time require to retrain the ranking model regularly with freshly collected data. This limits the time that is available for training to prohibitively short intervals. Third, ranking in real-time is necessary whenever the query complexity prevents us from caching the predictions. This constraint requires to minimize prediction time (or equivalently maximize the data throughput), which in turn may prevent us from achieving the accuracy necessary in web-scale industrial applications. In this paper, we investigate how the practical challenges faced in this setting can be tackled via an online learning to rank approach. Sparse models will be the key to reduce prediction latency, whereas one-pass stochastic optimization will minimize the training time and restrict the memory footprint. Interestingly, and perhaps surprisingly, extensive experiments show that one-pass learning preserves most of the predictive performance. Additionally, we study a variety of online learning algorithms that enforce sparsity and provide insights to help the practitioner make an informed decision about which approach to pick. We report results on a massive purchase log dataset from the Amazon retail website, as well as on several benchmarks from the LETOR corpus.

Joint semi-supervised learning of Hidden Conditional Random Fields and Hidden Markov Models

Although semi-supervised learning has generated great interest for designing classifiers on static patterns, there has been comparatively fewer works on semi-supervised learning for structured outputs and in particular for sequences. We investigate semi-supervised approaches for learning hidden state conditional random fields for sequence classification. We propose a new approach that iteratively learns a pair of discriminative-generative models, namely Hidden Markov Models (HMMs) and Hidden Conditional Random Fields (HCRFs). Our method builds on simple strategies for semi-supervised learning of HMMs and on strategies for initializing HCRFs from HMMs. We investigate the behavior of the method on artificial data and provide experimental results for two real problems, handwritten character recognition and financial chart pattern recognition. We compare our approach with state of the art semi-supervised methods.

The Geography of Online News Engagement

Geographical processes might well impact online engagement in big countries like the USA. Upon a random sample of 200K news articles and corresponding 41M comments posted on the Yahoo! News in that country, we show that nearby individuals tend to comment and engage with similar news articles more than distant individuals do. Interestingly, at state level, topics one reads about are associated with specific socio-economic conditions and personality traits.

Human Atlas: A Tool for Mapping Social Networks

Most social network analyses focus on online social networks. While these networks encode important aspects of our lives they fail to capture many real-world social connections. Most of these connections are, in fact, public and known to the members of the community. Mapping them is a task very suitable for crowdsourcing: it is easily broken down in many simple and independent subtasks. Due to the nature of social networks-presence of highly connected nodes and tightly knit groups-if we allow users to map their immediate connections and the connections between them, we will need few participants to map most connections within a community. To this end, we built the Human Atlas, a web-based tool for mapping social networks. To test it, we partially mapped the social network of the MIT Media Lab. We ran a user study and invited members of the community to use the tool. In 4.6 man-hours, 22 participants mapped 984 connections within the lab, demonstrating the potential of the tool.

Me, My Echo Chamber, and I: Introspection on Social Media Polarization

Homophily - our tendency to surround ourselves with others who share our perspectives and opinions about the world - is both a part of human nature and an organizing principle underpinning many of our digital social networks. However, when it comes to politics or culture, homophily can amplify tribal mindsets and produce echo chambers that degrade the quality, safety, and diversity of discourse online. While several studies have empirically proven this point, few have explored how making users aware of the extent and nature of their political echo chambers influences their subsequent beliefs and actions. In this paper, we introduce Social Mirror, a social network visualization tool that enables a sample of Twitter users to explore the politically-active parts of their social network. We use Social Mirror to recruit Twitter users with a prior history of political discourse to a randomized experiment where we evaluate the effects of different treatments on participants i) beliefs about their network connections, ii) the political diversity of who they choose to follow, and iii) the political alignment of the URLs they choose to share. While we see no effects on average political alignment of shared URLs, we find that recommending accounts of the opposite political ideology to follow reduces participants» beliefs in the political homogeneity of their network connections but still enhances their connection diversity one week after treatment. Conversely, participants who enhance their belief in the political homogeneity of their Twitter connections have less diverse network connections 2-3 weeks after treatment. We explore the implications of these disconnects between beliefs and actions on future efforts to promote healthier exchanges in our digital public spheres.

Light it up: using paper circuitry to enhance low-fidelity paper prototypes for children

Paper prototyping is an important tool for designing and testing early technologies during development. However, children have different relationships with technology and thus one cannot expect children to assess paper prototypes with the same mental model as adults. In this paper, we examine the effect of incorporating paper circuitry into low-fidelity paper prototypes, in order to add a level of interactivity that is not present in traditional paper prototypes. We conducted a study with 20 children ages 3 to 10 years old where participants used a cardboard prototype of a voice-controlled rocket on a pretend play mission to Mars. Children chose between buttons that lit up when pressed using paper circuitry, and buttons that did not light up, and explained their selections to the researchers. Our results show that children indeed preferred buttons augmented with paper circuitry, demonstrating more attention for and increased believability in the function of these buttons as well as the overall system. These findings show how designers can use paper circuitry to more effectively engage children while play-testing their paper prototypes.

Development of an English-Macedonian Machine Readable Dictionary by Using Parallel Corpora

The dictionaries are one of the most useful lexical resources. However, most of the dictionaries today are not in digital form. This makes them cumbersome for usage by humans and impossible for integration in computer programs. The process of digitalizing an existing traditional dictionary is expensive and labor intensive task. In this paper, we present a method for development of Machine Readable Dictionaries by using the already available resources. Machine readable dictionary consists of simple word-toword mappings, where word from the source language can be mapped into several optional words in the target language. We present a series of experiments where by using the parallel corpora and open source Statistical Machine Translation tools at our disposal, we managed to develop an English- Macedonian Machine Readable Dictionary containing 23,296 translation pairs (17,708 English and 18,343 Macedonian terms). A subset of the produced dictionary has been manually evaluated and showed accuracy of 79.8%.