Luca Chiarandini

Buon appetito: recommending personalized menus

This paper deals with the problem of menu recommendation, namely recommending menus that a person is likely to consume at a particular restaurant. We mine restaurant reviews to extract food words, we use sentiment analysis applied to each sentence in order to compute the individual food preferences. Then we extract frequent combination of dishes using a variation of the Apriori algorithm. Finally, we propose several recommender systems to provide suggestions of food items or entire menus, i.e. sets of dishes.

Discriminative clustering for market segmentation

We study discriminative clustering for market segmentation tasks. The underlying problem setting resembles discriminative clustering, however, existing approaches focus on the prediction of univariate cluster labels. By contrast, market segments encode complex (future) behavior of the individuals which cannot be represented by a single variable. In this paper, we generalize discriminative clustering to structured and complex output variables that can be represented as graphical models. We devise two novel methods to jointly learn the classifier and the clustering using alternating optimization and collapsed inference, respectively. The two approaches jointly learn a discriminative segmentation of the input space and a generative output prediction model for each segment. We evaluate our methods on segmenting user navigation sequences from Yahoo! News. The proposed collapsed algorithm is observed to outperform baseline approaches such as mixture of experts. We showcase exemplary projections of the resulting segments to display the interpretability of the solutions.

Search behaviour on photo sharing platforms

The behaviour, goals, and intentions of users while searching for images in large scale online collections are not well understood, with image search log analysis providing limited insights, in part because they tend only to have access to user search and result click information. In this paper we study user search behaviour in a large photo-sharing platform, analyzing all user actions during search sessions (i.e. including post result-click pageviews). Search accounts for a significant part of user interactions with such platforms, and we show differences between the queries issued on such platforms and those on general image search. We show that search behaviour is influenced by the query type, and also depends on the user. Finally, we analyse how users behave when they reformulate their queries, and develop URL class prediction models for image search, showing that query-specific models significantly outperform query-agnostic models. The insights provided in this paper are intended as a launching point for the design of better interfaces and ranking models for image search.

A system for dynamic playlist generation driven by multimodal control signals and descriptors

This work describes a general approach to multimedia playlist generation and description and an application of the approach to music information retrieval. The example of system that we implemented updates a musical playlist on the fly based on prior information (musical preferences); current descriptors of the song that is being played; and fine-grained and semantically rich descriptors (descriptors of users gestures, of environment conditions, etc.). The system incorporates a learning system that infers the users preferences. Subjective tests have been conducted on usability and quality of the recommendation system.

Exploration and discovery of user-generated content in large information spaces

The accumulation of large collections of social media data poses new challenges for the design of exploratory experiences, such as when a user browses through a collection to discover content (e.g. exploring photo collections, network of friends, etc). Cardinality and characteristics of the set, together with volatility of the information, resulting from fast and continuous creation, deletion and updating of entries, trigger novel research questions. In this context, we plan to investigate and contribute to the data analysis, and user interface design of exploratory experiences. The proposed approach is an iterative process where analysis and design phases are performed in cycles. The long-term vision is to understand the underlying reasoning in order to be able to automatically replicate it.

Metro: Exploring Participation in Public Events

The structure of a social network is time-dependent, as relationships between entities change in time. In large networks, static or animated visualizations are often insufficient to capture all the information about the interactions between people over time, which could be captured better by interactive interfaces. We propose a novel system for exploring the interactions of entities over time, and support it with an application that displays interactions of public figures at events.

PRiSMA: searching images in parallel

PRiSMA is an image search application for tablet and desktop devices intended to facilitate and promote the searching of images in parallel. With an intuitive user interface, users can branch their queries into multiple horizontal sliding strips to simultaneously explore different perspectives of large image collections (e.g., colors, geographical location or topic). Strips can be easily created, tailored, merged, and removed, allowing users to effectively perform multiple queries and manage the results in a dynamic and orderly fashion. With PRiSMA we aim to explore the potential and limitations of parallel image search from a user perspective.