Enrico Palumbo

entity2rec: Learning User-Item Relatedness from Knowledge Graphs for Top-N Item Recommendation

Knowledge Graphs have proven to be extremely valuable to recommender systems, as they enable hybrid graph-based recommendation models encompassing both collaborative and content information. Leveraging this wealth of heterogeneous information for top-N item recommendation is a challenging task, as it requires the ability of effectively encoding a diversity of semantic relations and connectivity patterns. In this work, we propose entity2rec, a novel approach to learning user-item relatedness from knowledge graphs for top-N item recommendation. We start from a knowledge graph modeling user-item and item-item relations and we learn property-specific vector representations of users and items applying neural language models on the network. These representations are used to create property-specific user-item relatedness features, which are in turn fed into learning to rank algorithms to learn a global relatedness model that optimizes top-N item recommendations. We evaluate the proposed approach in terms of ranking quality on the MovieLens 1M dataset, outperforming a number of state-of-the-art recommender systems, and we assess the importance of property-specific relatedness scores on the overall ranking quality.

3cixty: Building Comprehensive Knowledge Bases For City Exploration

Abstract Planning a visit to Expo Milano 2015 or simply touring in Milan are activities that require a certain amount of a priori knowledge of the city. In this paper, we present the process of building such comprehensive knowledge bases that contain descriptions of events and activities, places and sights, transportation facilities as well as social activities, collected from numerous static, near- and real-time local and global data providers, including hyper local sources such as the Expo Milano 2015 official services and several social media platforms. Entities in the 3cixty KB are deduplicated, interlinked and enriched using semantic technologies. The 3cixty KB is empowering the ExplorMI 360 multi-device application, which has been officially endorsed by the E015 Technical Management Board and has gained the patronage of Expo Milano in 2015, thus has offered a unique testing scenario for the 20 million visitors along the 6 months of the exhibit. In 2016–2017, new knowledge bases have been created for the cities of London, Madeira and Singapore, as well as for the entire French Cote d’Azur area. The 3cixty KB is accessible at https://kb.3cixty.com/sparql while ExplorMI 360 at https://www.3cixty.com and in the Google Play Store and Apple App Store.

An Ensemble Approach to Financial Entity Matching for the FEIII 2016 Challenge

Financial entities are often referred to with ambiguous descriptions and identifiers. To tackle this issue, the Financial Entity Identification and Information Integration1 (FEIII) Challenge requires participants to automatically reconcile financial entities among three datasets: the Federal Financial Institution Examination Council2 (FFIEC), the Legal Entity Identifiers (LEI) and the Security and Exchange Commission3 (SEC). Our approach is based on the combination of different Naive Bayes classifiers through an ensemble approach. The evaluation on the Gold Standard developed by the challenge organizers shows F1-scores that are above the average of the other participants for the two proposed tasks.

TwitPersonality: Computing Personality Traits from Tweets Using Word Embeddings and Supervised Learning

We are what we do, like, and say. Numerous research efforts have been pushed towards the automatic assessment of personality dimensions relying on a set of information gathered from social media platforms such as list of friends, interests of musics and movies, endorsements and likes an individual has ever performed. Turning this information into signals and giving them as inputs to supervised learning approaches has resulted in being particularly effective and accurate in computing personality traits and types. Despite the demonstrated accuracy of these approaches, the sheer amount of information needed to put in place such a methodology and access restrictions make them unfeasible to be used in a real usage scenario. In this paper, we propose a supervised learning approach to compute personality traits by only relying on what an individual tweets about publicly. The approach segments tweets in tokens, then it learns word vector representations as embeddings that are then used to feed a supervised learner classifier. We demonstrate the effectiveness of the approach by measuring the mean squared error of the learned model using an international benchmark of Facebook status updates. We also test the transfer learning predictive power of this model with an in-house built benchmark created by twenty four panelists who performed a state-of-the-art psychological survey and we observe a good conversion of the model while analyzing their Twitter posts towards the personality traits extracted from the survey.

Assessing the Impacts of Electric Vehicles Uptake: A System Dynamics Approach

The shift from conventional cars to Electric Vehicles (EVs) may significantly improve air quality and consequently public health, especially if electricity is powered by renewable energy. This paper describes a System Dynamics model to estimate the environmental health impacts of alternative market scenarios for EVs diffusion in Piedmont (Italy). The scenarios account for: (a) different market shares of Battery Electric Vehicles (BEVs) and Plug-in Hybrid Electric Vehicles (PHEVs), and (b) different trends of EVs uptake, in accordance with official electric mobility studies. The causal relationships among the variables of this complex system generate a positive feedback loop, involving: (1) EVs uptake, (2) pollutants concentration, (3) hospital admissions, and (4) public health savings. The simulation model results in an ex-ante impact evaluation tool, which is able to perform what-if analyses of different electric mobility scenarios and to compare their 2030 projections versus a reference scenario, thus allowing a counterfactual analysis able to isolate the impacts due solely to specific actions. This study has important policy implications, since the proposed model: (a) provides an integrated and comprehensive framework that allows to account for all the three dimensions of sustainability (economic, environmental, social), (b) gives ex-ante estimates of environmental health impacts of EVs uptake according to different scenarios, and quantifies the savings derived by the reduction of public health spending, converting them in monetary incentives for EVs purchase, (c) proposes recommendations for the definition of Sustainable Urban Mobility Plans of a smart city.

Semantic Similarity Between Images: A Novel Approach Based on a Complex Network of Free Word Associations

Several measures exist to describe similarities between digital contents, especially for what concerns images. Nevertheless, distances based on low-level visual features embedded in a multidimensional linear space are hardly suitable for capturing semantic similarities and recently novel techniques have been introduced making use of hierarchical knowledge bases. While being successfully exploited in specific contexts, the human perception of similarity cannot be easily encoded in such rigid structures. In this paper we propose to represent a knowledge base of semantic concepts as a complex network whose topology arises from free conceptual associations and is markedly different from a hierarchical structure. Images are anchored to relevant semantic concepts through an annotation process and similarity is computed following the related paths in the complex network. We finally show how this definition of semantic similarity is not necessarily restricted to images, but can be extended to compute distances between different types of sensorial information such as pictures and sounds, modeling the human ability to realize synaesthesias.

A complex network model of semantic memory impairments

In the last decades, several models have been proposed to describe the functions and the structure of human memory. Many of these agree in representing semantic memory, i.e. the part of memory which contains the general knowledge about the world, as a network. On the other hand, the study of complex networks is a new and emerging field at the intersection of physics, mathematics and computer science which aims at characterizing the topological properties of large networks. The paper proposes a quantitative study of the large-scale properties of semantic memory, modelled as the knowledge base of an automatic concept classifier of images. This approach allows us to probe the topological properties of the network, showing that it exhibits the marks of complexity, and provide us with a suitable mathematical framework to study memory impairments. These alterations are firstly modelled as nodes removals and secondly as links modifications, producing markedly different results.