Sanmitra Bhattacharya

The gene normalization task in BioCreative III

BackgroundWe report the Gene Normalization (GN) challenge in BioCreative III where participating teams were asked to return a ranked list of identifiers of the genes detected in full-text articles. For training, 32 fully and 500 partially annotated articles were prepared. A total of 507 articles were selected as the test set. Due to the high annotation cost, it was not feasible to obtain gold-standard human annotations for all test articles. Instead, we developed an Expectation Maximization (EM) algorithm approach for choosing a small number of test articles for manual annotation that were most capable of differentiating team performance. Moreover, the same algorithm was subsequently used for inferring ground truth based solely on team submissions. We report team performance on both gold standard and inferred ground truth using a newly proposed metric called Threshold Average Precision (TAP-k).ResultsWe received a total of 37 runs from 14 different teams for the task. When evaluated using the gold-standard annotations of the 50 articles, the highest TAP-k scores were 0.3297 (k=5), 0.3538 (k=10), and 0.3535 (k=20), respectively. Higher TAP-k scores of 0.4916 (k=5, 10, 20) were observed when evaluated using the inferred ground truth over the full test set. When combining team results using machine learning, the best composite system achieved TAP-k scores of 0.3707 (k=5), 0.4311 (k=10), and 0.4477 (k=20) on the gold standard, representing improvements of 12.4%, 21.8%, and 26.6% over the best team results, respectively.ConclusionsBy using full text and being species non-specific, the GN task in BioCreative III has moved closer to a real literature curation task than similar tasks in the past and presents additional challenges for the text mining community, as revealed in the overall team results. By evaluating teams using the gold standard, we show that the EM algorithm allows team submissions to be differentiated while keeping the manual annotation effort feasible. Using the inferred ground truth we show measures of comparative performance between teams. Finally, by comparing team rankings on gold standard vs. inferred ground truth, we further demonstrate that the inferred ground truth is as effective as the gold standard for detecting good team performance.

BioCreative III interactive task: an overview

BackgroundThe BioCreative challenge evaluation is a community-wide effort for evaluating text mining and information extraction systems applied to the biological domain. The biocurator community, as an active user of biomedical literature, provides a diverse and engaged end user group for text mining tools. Earlier BioCreative challenges involved many text mining teams in developing basic capabilities relevant to biological curation, but they did not address the issues of system usage, insertion into the workflow and adoption by curators. Thus in BioCreative III (BC-III), the InterActive Task (IAT) was introduced to address the utility and usability of text mining tools for real-life biocuration tasks. To support the aims of the IAT in BC-III, involvement of both developers and end users was solicited, and the development of a user interface to address the tasks interactively was requested.ResultsA User Advisory Group (UAG) actively participated in the IAT design and assessment. The task focused on gene normalization (identifying gene mentions in the article and linking these genes to standard database identifiers), gene ranking based on the overall importance of each gene mentioned in the article, and gene-oriented document retrieval (identifying full text papers relevant to a selected gene). Six systems participated and all processed and displayed the same set of articles. The articles were selected based on content known to be problematic for curation, such as ambiguity of gene names, coverage of multiple genes and species, or introduction of a new gene name. Members of the UAG curated three articles for training and assessment purposes, and each member was assigned a system to review. A questionnaire related to the interface usability and task performance (as measured by precision and recall) was answered after systems were used to curate articles. Although the limited number of articles analyzed and users involved in the IAT experiment precluded rigorous quantitative analysis of the results, a qualitative analysis provided valuable insight into some of the problems encountered by users when using the systems. The overall assessment indicates that the system usability features appealed to most users, but the system performance was suboptimal (mainly due to low accuracy in gene normalization). Some of the issues included failure of species identification and gene name ambiguity in the gene normalization task leading to an extensive list of gene identifiers to review, which, in some cases, did not contain the relevant genes. The document retrieval suffered from the same shortfalls. The UAG favored achieving high performance (measured by precision and recall), but strongly recommended the addition of features that facilitate the identification of correct gene and its identifier, such as contextual information to assist in disambiguation.DiscussionThe IAT was an informative exercise that advanced the dialog between curators and developers and increased the appreciation of challenges faced by each group. A major conclusion was that the intended users should be actively involved in every phase of software development, and this will be strongly encouraged in future tasks. The IAT Task provides the first steps toward the definition of metrics and functional requirements that are necessary for designing a formal evaluation of interactive curation systems in the BioCreative IV challenge.

Engagement with Health Agencies on Twitter

Objective To investigate factors associated with engagement of U.S. Federal Health Agencies via Twitter. Our specific goals are to study factors related to a) numbers of retweets, b) time between the agency tweet and first retweet and c) time between the agency tweet and last retweet. Methods We collect 164,104 tweets from 25 Federal Health Agencies and their 130 accounts. We use negative binomial hurdle regression models and Cox proportional hazards models to explore the influence of 26 factors on agency engagement. Account features include network centrality, tweet count, numbers of friends, followers, and favorites. Tweet features include age, the use of hashtags, user-mentions, URLs, sentiment measured using Sentistrength, and tweet content represented by fifteen semantic groups. Results A third of the tweets (53,556) had zero retweets. Less than 1% (613) had more than 100 retweets (mean  = 284). The hurdle analysis shows that hashtags, URLs and user-mentions are positively associated with retweets; sentiment has no association with retweets; and tweet count has a negative association with retweets. Almost all semantic groups, except for geographic areas, occupations and organizations, are positively associated with retweeting. The survival analyses indicate that engagement is positively associated with tweet age and the follower count. Conclusions Some of the factors associated with higher levels of Twitter engagement cannot be changed by the agencies, but others can be modified (e.g., use of hashtags, URLs). Our findings provide the background for future controlled experiments to increase public health engagement via Twitter.

MeSH: a window into full text for document summarization

Motivation: Previous research in the biomedical text-mining domain has historically been limited to titles, abstracts and metadata available in MEDLINE records. Recent research initiatives such as TREC Genomics and BioCreAtIvE strongly point to the merits of moving beyond abstracts and into the realm of full texts. Full texts are, however, more expensive to process not only in terms of resources needed but also in terms of accuracy. Since full texts contain embellishments that elaborate, contextualize, contrast, supplement, etc., there is greater risk for false positives. Motivated by this, we explore an approach that offers a compromise between the extremes of abstracts and full texts. Specifically, we create reduced versions of full text documents that contain only important portions. In the long-term, our goal is to explore the use of such summaries for functions such as document retrieval and information extraction. Here, we focus on designing summarization strategies. In particular, we explore the use of MeSH terms, manually assigned to documents by trained annotators, as clues to select important text segments from the full text documents. Results: Our experiments confirm the ability of our approach to pick the important text portions. Using the ROUGE measures for evaluation, we were able to achieve maximum ROUGE-1, ROUGE-2 and ROUGE-SU4 F-scores of 0.4150, 0.1435 and 0.1782, respectively, for our MeSH term-based method versus the maximum baseline scores of 0.3815, 0.1353 and 0.1428, respectively. Using a MeSH profile-based strategy, we were able to achieve maximum ROUGE F-scores of 0.4320, 0.1497 and 0.1887, respectively. Human evaluation of the baselines and our proposed strategies further corroborates the ability of our method to select important sentences from the full texts. Contact: sanmitra-bhattacharya@uiowa.edu; padmini-srinivasan@uiowa.edu

Belief surveillance with Twitter

Data from social media systems are being actively mined for trends and patterns of interests. Problems such as sentiment and opinion mining, and prediction of election outcomes have become tremendously popular due to the unprecedented availability of social interactivity data of different types. An important angle that has not yet been explored is to estimate beliefs from posts made on social media. We propose that social media can be used to monitor the level of belief, disbelief and doubt related to specific propositions. Inspired by efforts in disease surveillance using social media we coin the term belief surveillance for this function. We propose a novel methodological framework for belief surveillance using Twitter. Our method may be used to gauge belief on any proposition as long as it is specifiable in a form that we call probes. We present our belief estimates for 32 probes some of which represent factual information, others represent false information and the remaining represent debatable propositions. Finally, we provide preliminary evidence suggesting that off-the-shelf classifiers may be used to automatically estimate belief.

Perceptions of presidential candidates' personalities in twitter

Political sentiment analysis using social media, especially Twitter, has attracted wide interest in recent years. In such research, opinions about politicians are typically divided into positive, negative, or neutral. In our research, the goal is to mine political opinion from social media at a higher resolution by assessing statements of opinion related to the personality traits of politicians; this is an angle that has not yet been considered in social media research. A second goal is to contribute a novel retrieval‐based approach for tracking public perception of personality using Gough and Heilbruns Adjective Check List (ACL) of 110 terms describing key traits. This is in contrast to the typical lexical and machine‐learning approaches used in sentiment analysis. High‐precision search templates developed from the ACL were run on an 18‐month span of Twitter posts mentioning Obama and Romney and these retrieved more than half a million tweets. For example, the results indicated that Romney was perceived as more of an achiever and Obama was perceived as somewhat more friendly. The traits were also aggregated into 14 broad personality dimensions. For example, Obama rated far higher than Romney on the Moderation dimension and lower on the Machiavellianism dimension. The temporal variability of such perceptions was explored.

Social media engagement analysis of U.S. Federal health agencies on Facebook

BackgroundIt is becoming increasingly common for individuals and organizations to use social media platforms such as Facebook. These are being used for a wide variety of purposes including disseminating, discussing and seeking health related information. U.S. Federal health agencies are leveraging these platforms to ‘engage’ social media users to read, spread, promote and encourage health related discussions. However, different agencies and their communications get varying levels of engagement. In this study we use statistical models to identify factors that associate with engagement.MethodsWe analyze over 45,000 Facebook posts from 72 Facebook accounts belonging to 24 health agencies. Account usage, user activity, sentiment and content of these posts are studied. We use the hurdle regression model to identify factors associated with the level of engagement and Cox proportional hazards model to identify factors associated with duration of engagement.ResultsIn our analysis we find that agencies and accounts vary widely in their usage of social media and activity they generate. Statistical analysis shows, for instance, that Facebook posts with more visual cues such as photos or videos or those which express positive sentiment generate more engagement. We further find that posts on certain topics such as occupation or organizations negatively affect the duration of engagement.ConclusionsWe present the first comprehensive analyses of engagement with U.S. Federal health agencies on Facebook. In addition, we briefly compare and contrast findings from this study to our earlier study with similar focus but on Twitter to show the robustness of our methods.

Data-Driven Methods for SMS-Based FAQ Retrieval

SMS text messaging is one of the most popular data applications on mobile phones these days. Other than personal communication, text messaging can also be used for various purposes like bill payment, banking, inquiry, etc. However these messages are extremely noisy and contain misspellings, abbreviations, transliterations, etc. Keeping this in mind, FIRE 2011 introduced a new retrieval task called SMS-based FAQ retrieval in English, Hindi and Malayalam. Within-language and cross-language tasks were designed for this retrieval problem. As solutions we propose various data-driven retrieval techniques that includes noise reduction in the SMS queries and the FAQ corpora. Overall, we find that our methods work well for the retrieval experiments in the different languages. For English, the use of Google Spelling Suggestions and term expansion strategies improve retrieval scores. For Hindi and Malayalam retrieval experiments, we find that translation of queries and corpus to English increases retrieval accuracy.

A semantic approach to involve Twitter in LBD efforts

Literature-based Discovery (LBD) refers to the task of finding hidden, unknown or neglected relationships that may be uncovered using biomedical text. While traditional LBD primarily focuses on MEDLINE records for unearthing such relationships, recent studies have shown the applicability of contemporary textual resources such as electronic medical records or online medical message boards for similar purposes. In this paper we highlight yet another source for LBD, i.e., Twitter data. We focus on the use of Twitter as a new resource for finding hypotheses - both novel and slightly studied. Using a set of drug and disease names as starting points we retrieve thousands of Twitter messages which are then processed for semantic information to mine several hundred biomedical relationships which we call probes. Manual inspection of a handful of these probes reveals instances where tweets strongly support a hypothesis for which no evidence can be found in PubMed. In other cases, we find very few related PubMed records supporting/rejecting such Twitter-mined probes. Overall, we show the importance and usefulness of Twitter for LBD efforts.