Giovanni Stilo

Influenza-Like Illness Surveillance on Twitter through Automated Learning of Naïve Language

Twitter has the potential to be a timely and cost-effective source of data for syndromic surveillance. When speaking of an illness, Twitter users often report a combination of symptoms, rather than a suspected or final diagnosis, using naïve, everyday language. We developed a minimally trained algorithm that exploits the abundance of health-related web pages to identify all jargon expressions related to a specific technical term. We then translated an influenza case definition into a Boolean query, each symptom being described by a technical term and all related jargon expressions, as identified by the algorithm. Subsequently, we monitored all tweets that reported a combination of symptoms satisfying the case definition query. In order to geolocalize messages, we defined 3 localization strategies based on codes associated with each tweet. We found a high correlation coefficient between the trend of our influenza-positive tweets and ILI trends identified by US traditional surveillance systems.

Temporal Semantics: Time-Varying Hashtag Sense Clustering

Hashtags are creative labels used in micro-blogs to characterize the topic of a message/discussion. However, since hashtags are created in a spontaneous and highly dynamic way by users using multiple languages, the same topic can be associated to different hashtags and conversely, the same hashtag may imply different topics in different time spans. Contrary to common words, sense clustering for hashtags is complicated by the fact that no sense catalogues are available, like, e.g. Wikipedia or WordNet and furthermore, hashtag labels are often obscure. In this paper we propose a sense clustering algorithm based on temporal mining. First, hashtag time series are converted into strings of symbols using Symbolic Aggregate ApproXimation (SAX), then, hashtags are clustered based on string similarity and temporal co-occurrence. Evaluation is performed on two reference datasets of semantically tagged hashtags. We also perform a complexity evaluation of our algorithm, since efficiency is a crucial performance factor when processing large-scale data streams, such as Twitter.

Can Twitter Be a Source of Information on Allergy? Correlation of Pollen Counts with Tweets Reporting Symptoms of Allergic Rhinoconjunctivitis and Names of Antihistamine Drugs

Pollen forecasts are in use everywhere to inform therapeutic decisions for patients with allergic rhinoconjunctivitis (ARC). We exploited data derived from Twitter in order to identify tweets reporting a combination of symptoms consistent with a case definition of ARC and those reporting the name of an antihistamine drug. In order to increase the sensitivity of the system, we applied an algorithm aimed at automatically identifying jargon expressions related to medical terms. We compared weekly Twitter trends with National Allergy Bureau weekly pollen counts derived from US stations, and found a high correlation of the sum of the total pollen counts from each stations with tweets reporting ARC symptoms (Pearson’s correlation coefficient: 0.95) and with tweets reporting antihistamine drug names (Pearson’s correlation coefficient: 0.93). Longitude and latitude of the pollen stations affected the strength of the correlation. Twitter and other social networks may play a role in allergic disease surveillance and in signaling drug consumptions trends.

Recommendation of microblog users based on hierarchical interest profiles

Quite a number of recent works have concentrated on the task of recommending to Twitter users whom they should follow, among which, the WTF (Who To Follow) service provided by Twitter. Recommenders are based, either on the user’s network structure, or on some notion of topical similarity with other users, or on both. In this paper, we propose to accomplish the recommendation task in two steps: First, we profile users and classify them as belonging to a target community (depending e.g., on their political affiliation, preferred football team, favorite coffee shop, etc.). Then, we fine-tune recommendations for selected populations. We cast both problems of user classification and recommendation as one of itemset mining, where items are either users’ authoritative friends or semantic categories associated to friends, extracted from WiBi, the Wikipedia Bitaxonomy. In addition to evaluating our profiler and recommender on several populations, we also show that semantic categories allow for very fine-grained population studies, and make it possible to recommend not only whom to follow, but also topics of interest, users interested in the same topic, and more.

Hashtag sense clustering based on temporal similarity

Hashtags are creative labels used in micro-blogs to characterize the topic of a message/discussion. Regardless of the use for which they were originally intended, hashtags cannot be used as a means to cluster messages with similar content. First, because hashtags are created in a spontaneous and highly dynamic way by users in multiple languages, the same topic can be associated with different hashtags, and conversely, the same hashtag may refer to different topics in different time periods. Second, contrary to common words, hashtag disambiguation is complicated by the fact that no sense catalogs (e.g., Wikipedia or WordNet) are available; and, furthermore, hashtag labels are difficult to analyze, as they often consist of acronyms, concatenated words, and so forth. A common way to determine the meaning of hashtags has been to analyze their context, but, as we have just pointed out, hashtags can have multiple and variable meanings. In this article, we propose a temporal sense clustering algorithm based on the idea that semantically related hashtags have similar and synchronous usage patterns.

Predicting Flu Epidemics Using Twitter and Historical Data

Recently there has been a growing attention on the use of web and social data to improve traditional prediction models in politics, finance, marketing and health, but even though a correlation between observed phenomena and related social data has been demonstrated in many cases, yet the effectiveness of the latter for long-term or even mid-term predictions has not been shown. In epidemiological surveillance, the problem is compounded by the fact that infectious diseases models (such as susceptible-infected-recovered-susceptible, SIRS) are very sensitive to current conditions, such that small changes can produce remarkable differences in future outcomes. Unfortunately, current or nearly-current conditions keep changing as data are collected and updated by the epidemiological surveillance organizations. In this paper we show that the time series of Twitter messages reporting a combination of symptoms that match the influenza-like-illness (ILI) case definition represent a more stable and reliable information on “current conditions”, to the point that they can replace, rather than simply integrate, official epidemiological data. We estimate the effectiveness of these data at predicting current and past flu seasons (17 seasons overall), in combination with official historical data on past seasons, obtaining an average correlation of 0.85 over a period of 17 weeks covering the flu season.

Automatic Acquisition of a Taxonomy of Microblogs Users’ Interests

Abstract Modeling users’ interests plays an important role in the current web since it is at the basis of many services such as recommendation and customization. Using semantic technologies to represent users’ interests may help to reduce problems such as sparsity, over-specialization and domain-dependency, which are known to be critical issues of state of the art recommenders. In this paper we present a method for high-coverage modeling of Twitter users supported by a hierarchical representation of their interests, which we call a Twixonomy. In order to automatically build a population, community, or single-user Twixonomy we first identify “topical” friends in users’ friendship lists (i.e., friends representing an interest rather than a social relation between peers). We classify as topical those users with an associated page on Wikipedia. A word-sense disambiguation algorithm is used to select the appropriate Wikipedia page for each topical friend. Next, starting from the set of wikipages representing the main topics of interests of the considered Twitter population, we extract all paths connecting these pages with topmost Wikipedia category nodes, and we then prune the resulting graph efficiently so as to induce a direct acyclic graph and significantly reduce over ambiguity, a well known problem of the Wikipedia category graph. We release the Twixonomy produced in this work under creative common license.

A topic recommender for journalists

The way in which people gather information about events and form their own opinion on them has changed dramatically with the advent of social media. For many readers, the news gathered from online sources has become an opportunity to share points of view and information within micro-blogging platforms such as Twitter, mainly aimed at satisfying their communication needs. Furthermore, the need to deepen the aspects related to news stimulates a demand for additional information which is often met through online encyclopedias, such as Wikipedia. This behaviour has also influenced the way in which journalists write their articles, requiring a careful assessment of what actually interests the readers. The goal of this paper is to present a recommender system, What to Write and Why, capable of suggesting to a journalist, for a given event, the aspects still uncovered in news articles on which the readers focus their interest. The basic idea is to characterize an event according to the echo it receives in online news sources and associate it with the corresponding readers’ communicative and informative patterns, detected through the analysis of Twitter and Wikipedia, respectively. Our methodology temporally aligns the results of this analysis and recommends the concepts that emerge as topics of interest from Twitter and Wikipedia, either not covered or poorly covered in the published news articles.

Detecting network leaders in enterprises

This paper describes an interdisciplinary study aimed at analyzing leadership in less formal collaboration environments, such as enterprise social networks (ESNs). To conduct our research, we defined a measure of network leadership which draws on organization theory and on a computational model based on multiplex networks. This model, along with a social network analysis toolkit developed in the context of the present study, enabled the systematic empirical analysis of a large ESN, as a function of gender, time, roles, and discussed topics.