Search

The emergence of online social networks (OSNs) and the accompanying availability of large amounts of data, pose a number of new natural language processing (NLP) and computational challenges. Data from OSNs is different to data from traditional sources (e.g. newswire). The texts are short, noisy and conversational. Another important issue is that data occurs in a real-time streams, needing immediate analysis that is grounded in time and context. In this paper we describe a new open-source framework for efficient text processing of streaming OSN data (available at www.trendminer-project.eu). Whilst researchers have made progress in adapting or creating text analysis tools for OSN data, a system to unify these tasks has yet to be built. Our system is focused on a real world scenario where fast processing and accuracy is paramount. We use the MapReduce framework for distributed computing and present running times for our system in order to show that scaling to online scenarios is feasible.We describe the components of the system and evaluate their accuracy. Our system supports easy integration of future modules in order to extend its functionality.

BACKGROUND: in hospitalised older adults, cognitive impairments are common and may be associated with functional outcomes. Our aim was to systematically review this association. METHOD: we systematically searched MEDLINE, CINAHL, AMED and PsycINFO from inception to April 2016. Non-English language studies were filtered out at search stage. All types of studies were considered for inclusion except reviews, conference abstracts, dissertations and case studies. Population: community-dwelling or institutionalised older adults aged 65 years or more, who are acutely hospitalised and have information on history of dementia and/or cognitive scores on admission. Setting: acute hospital (excluding critical care and subacute or intermediate care). Outcome of interest: change in a measure of physical function or disability between pre-admission or admission, and discharge or post-discharge. This review was registered on PROSPERO (CRD42016035978). RESULTS: the search returned 5,988 unique articles, of which 34 met inclusion criteria. All studies were observational, with 30 prospective and 4 retrospective from 14 countries, recruiting from general medicine (n = 11), geriatric medicine (n = 11) and mixed (n = 12) wards. Twenty-six studies (54,637 participants) were suitable for the quantitative synthesis. The meta-analysis suggested that cognitive impairment was associated with functional decline in hospitalised older adults (risk ratio (RR): 1.64; 95% confidence interval (CI): 1.45–1.86; P < 0.01). Results were similar in subanalyses focusing on diagnosis of dementia (RR: 1.36; 95% CI: 1.05–1.76; P = 0.02; n= 2,248) or delirium (RR: 1.55; 95% CI: 1.31–1.83; P < 0.01; n= 1,677). CONCLUSION: cognitive impairments seem associated with functional decline in hospitalised older people. Causality cannot be inferred, and limitations include low quality of studies and possible confounding.

Ontologies that evolve through use to support new domain tasks can grow extremely large. Moreover, large ontologies require more resources to use and have slower response times than small ones. To help address this problem, we present an on-line semantic forgetting algorithm that removes ontology fragments containing infrequently used or cheap to relearn concepts. We situate our algorithm in an extension of the widely used RoboCup Rescue platform, which provides simulated tasks to agents. We show that our agents send fewer messages and complete more tasks, and thus achieve a greater degree of success, than other state-of-the-art approaches.