Web Data Extraction from Query Result Pages Based on Visual and Content Features

A rapidly increasing number of Web databases are now become accessible via
their HTML form-based query interfaces. Query result pages are dynamically generated
in response to user queries, which encode structured data and are displayed for human
use. Query result pages usually contain other types of information in addition to query
results, e.g., advertisements, navigation bar etc. The problem of extracting structured data
from query result pages is critical for web data integration applications, such as comparison
shopping, meta-search engines etc, and has been intensively studied. A number of approaches
have been proposed. As the structures of Web pages become more and more complex, the
existing approaches start to fail, and most of them do not remove irrelevant contents which
may a®ect the accuracy of data record extraction. We propose an automated approach for
Web data extraction. First, it makes use of visual features and query terms to identify data
sections and extracts data records in these sections. We also represent several content and
visual features of visual blocks in a data section, and use them to ¯lter out noisy blocks.
Second, it measures similarity between data items in di®erent data records based on their
visual and content features, and aligns them into di®erent groups so that the data in the
same group have the same semantics. The results of our experiments with a large set of
Web query result pages in di®erent domains show that our proposed approaches are highly
e®ective.

Diagnostic accuracy and clinical management by realtime teledermatology. Results from the Northern Ireland arms of the UK Multicentre Teledermatology Trial

Diagnostic accuracy and management recommendations of realtime teledermatology consultations using low-cost telemedicine equipment were evaluated. Patients were seen by a dermatologist over a video-link and a diagnosis and treatment plan were recorded. This was followed by a face-to-face consultation on the same day to confirm the earlier diagnosis and management plan. A total of 351 patients with 427 diagnoses participated. Sixty-seven per cent of the diagnoses made over the video-link agreed with the face-to-face diagnosis. Clinical management plans were recorded for 214 patients with 252 diagnoses. For this cohort, 44% of the patients were seen by the same dermatologist at both consultations, while 56% were seen by a different dermatologist. In 64% of cases the same management plan was recommended at both consultations; a sub-optimum treatment plan was recommended in 8% of cases; and in 9% of cases the video-link management plans were judged to be inappropriate. In 20% of cases the dermatologist was unable to recommend a suitable management plan by video-link. There were significant differences in the ability to recommend an optimum management plan by video-link when a different dermatologist made the reference management plan. The results indicate that a high proportion of dermatological conditions can be successfully managed by realtime teledermatology.

Preliminary results from the Northern Ireland arms of the UK Multicentre Teledermatology Trial:is clinical management by realtime teledermatology possible?

Results from phase 1 of the UK Multicentre Teledermatology Trial demonstrated the diagnostic accuracy of realtime teledermatology using low-cost equipment. Phase 2 of the trial aimed to assess its effectiveness as a management tool for dermatological disease. Teledermatology consultations were organized between two health centres and two hospitals in Northern Ireland using low-cost videoconferencing equipment. For 205 patients seen by a dermatologist over the video-link a diagnosis and management plan were recorded. A subsequent face-to-face consultation was arranged on the same day to confirm the diagnosis and treatment regime. A comparison of these management plans revealed that the same plan was recommended in 64% of cases; the teledermatologist was unable to advocate a suitable management plan in 19% of cases; a suboptimal treatment plan was suggested by the teledermatologist in 6% of cases; and in 11% of cases, the teledermatologist suggested an inappropriate treatment plan. These findings indicate that appropriate clinical management was possible in approximately two-thirds of dermatology consultations via the video-link.

Patient satisfaction with realtime teledermatology in Northern Ireland

Teledermatology consultations were organized between two health centers and two hospitals in Northern Ireland using low-cost videoconferencing equipment. A prospective study of patient satisfaction was carried out. Following each teleconsultation, patients were asked to complete a questionnaire assessing their satisfaction with the service. Over 22 months, 334 patients were seen by a dermatologist over the video-link, and 292 patients (87%) completed the 16-item questionnaire. Patients reported universal satisfaction with the technical aspects of teledermatology. The quality of both the audio and the display was highly acceptable to patients. Personal experiences of the teledermatology consultation were also favourable: 85% felt comfortable using the video-link. The benefits of teledermatology were generally recognized: 88% of patients thought that a teleconsultation could save time. Patients found the teledermatology consultation to be as acceptable as the conventional dermatology consultation. These findings suggest overall patient satisfaction with realtime teledermatology.

Hierarchical Bayesian models in ecology: Reconstructing species interaction networks from non-homogeneous species abundance data

The relationships among organisms and their surroundings can be of immense complexity. To describe and understand an ecosystem as a tangled bank, multiple ways of interaction and their effects have to be considered, such as predation, competition, mutualism and facilitation. Understanding the resulting interaction networks is a challenge in changing environments, e.g. to predict knock-on effects of invasive species and to understand how climate change impacts biodiversity. The elucidation of complex ecological systems with their interactions will benefit enormously from the development of new machine learning tools that aim to infer the structure of interaction networks from field data. In the present study, we propose a novel Bayesian regression and multiple changepoint model (BRAM) for reconstructing species interaction networks from observed species distributions. The model has been devised to allow robust inference in the presence of spatial autocorrelation and distributional heterogeneity. We have evaluated the model on simulated data that combines a trophic niche model with a stochastic population model on a 2-dimensional lattice, and we have compared the performance of our model with L1-penalized sparse regression (LASSO) and non-linear Bayesian networks with the BDe scoring scheme. In addition, we have applied our method to plant ground coverage data from the western shore of the Outer Hebrides with the objective to infer the ecological interactions. (C) 2012 Elsevier B.V. All rights reserved.

The physical activity loyalty card scheme:Development and application of a novel system for incentivizing behaviour change

In the Public Health White Paper "Healthy Lives, Healthy People" (2010), the UK Government emphasised using incentives and "nudging" to encourage positive, healthy behaviour changes. However, there is little evidence that nudging is effective, in particular for increasing physical activity. We have created a platform to research the effectiveness of health-related behaviour change interventions and incentive schemes. The system consists of an outward-facing website, incorporating tools for incentivizing behaviour change, and a novel physical activity monitoring system. The monitoring system consists of the "Physical Activity Loyalty Card", which contains a passive RFID tag, and a contactless sensor network to detect the cards. This paper describes the application of this novel web-based system to investigate the effectiveness of non-cash incentives to "nudge" adults to undertake more physical activity. © 2012 ICST Institute for Computer Science, Social Informatics and Telecommunications Engineering.

Personalized medicine: Has it started yet? A reconstruction of the early history

Within the last few years the field personalized medicine entered the stage. Accompanied with great hopes and expectations it is believed that this field may have the potential to revolutionize medical and clinical care by utilizing genomics information about the individual patients themselves. In this paper, we reconstruct the early footprints of personalized medicine as reflected by information retrieved from PubMed and Google Scholar. That means we are providing a data-driven perspective of this field to estimate its current status and potential problems.

Glial and neuronal dysfunction in streptozotocin-induced diabetic rats

Neuronal dysfunction has been noted very soon after the induction of diabetes by streptozotocin injection in rats. It is not clear from anatomical evidence whether glial cell dysfunction accompanies the well-documented neuronal deficit. Here, we isolate the Müller cell driven slow-P3 component of the full-field electroretinogram and show that it is attenuated at 4 weeks following the onset of streptozotocin-hyperglycaemia. We also found a concurrent reduction in the sensitivity of the phototransduction cascade, as well as in the components of the electroretinogram known to indicate retinal ganglion cell and amacrine cell integrity. Our data support the idea that neuronal and Müller cell dysfunction occurs at the same time in streptozotocin-induced hyperglycaemia.

TProf: An energy profiler for task-parallel programs

We present TProf, an energy profiling tool for OpenMP-like task-parallel programs. To compute the energy consumed by each task in a parallel application, TProf dynamically traces the parallel execution and uses a novel technique to estimate the per-task energy consumption. To achieve this estimation, TProf apportions the total processor energy among cores and overcomes the limitation of current works which would otherwise make parallel accounting impossible to achieve. We demonstrate the value of TProf by characterizing a set of task parallel programs, where we find that data locality, memory access patterns and task working sets are responsible for significant variance in energy consumption between seemingly homogeneous tasks. In addition, we identify opportunities for fine-grain energy optimization by applying per-task Dynamic Voltage and Frequency Scaling (DVFS).

Machine Learning for Predictive Maintenance: A Multiple Classifiers Approach

In this paper a multiple classifier machine learning methodology for Predictive Maintenance (PdM) is presented. PdM is a prominent strategy for dealing with maintenance issues given the increasing need to minimize downtime and associated costs. One of the challenges with PdM is generating so called ’health factors’ or quantitative indicators of the status of a system associated with a given maintenance issue, and determining their relationship to operating costs and failure risk. The proposed PdM methodology allows dynamical decision rules to be adopted for maintenance management and can be used with high-dimensional and censored data problems. This is achieved by training multiple classification modules with different prediction horizons to provide different performance trade-offs in terms of frequency of unexpected breaks and unexploited lifetime and then employing this information in an operating cost based maintenance decision system to minimise expected costs. The effectiveness of the methodology is demonstrated using a simulated example and a benchmark semiconductor manufacturing maintenance problem.