Automatic keyphrase extraction: a comparison of methods

There are many published methods available for creating keyphrases for documents. Previous work in the field has shown that in a significant proportion of cases author selected keyphrases are not appropriate for the document they accompany. This requires the use of such automated methods to improve the use of keyphrases. Often the keyphrases are not updated when the focus of a paper changes or include keyphrases that are more classificatory than explanatory. The published methods are all evaluated using different corpora, typically one relevant to their field of study. This not only makes it difficult to incorporate the useful elements of algorithms in future work but also makes comparing the results of each method inefficient and ineffective. This paper describes the work undertaken to compare five methods across a common baseline of six corpora. The methods chosen were term frequency, inverse document frequency, the C-Value, the NC-Value, and a synonym based approach. These methods were compared to evaluate performance and quality of results, and to provide a future benchmark. It is shown that, with the comparison metric used for this study Term Frequency and Inverse Document Frequency were the best algorithms, with the synonym based approach following them. Further work in the area is required to determine an appropriate (or more appropriate) comparison metric.

A comparison of automated keyphrase extraction techniques and of automatic evaluation vs. human evaluation

Keyphrases are added to documents to help identify the areas of interest they contain. However, in a significant proportion of papers author selected keyphrases are not appropriate for the document they accompany: for instance, they can be classificatory rather than explanatory, or they are not updated when the focus of the paper changes. As such, automated methods for improving the use of keyphrases are needed, and various methods have been published. However, each method was evaluated using a different corpus, typically one relevant to the field of study of the method’s authors. This not only makes it difficult to incorporate the useful elements of algorithms in future work, but also makes comparing the results of each method inefficient and ineffective. This paper describes the work undertaken to compare five methods across a common baseline of corpora. The methods chosen were Term Frequency, Inverse Document Frequency, the C-Value, the NC-Value, and a Synonym based approach. These methods were analysed to evaluate performance and quality of results, and to provide a future benchmark. It is shown that Term Frequency and Inverse Document Frequency were the best algorithms, with the Synonym approach following them. Following these findings, a study was undertaken into the value of using human evaluators to judge the outputs. The Synonym method was compared to the original author keyphrases of the Reuters’ News Corpus. The findings show that authors of Reuters’ news articles provide good keyphrases but that more often than not they do not provide any keyphrases.

Studying the Human Gut Microbiota in the Trans-Omics Era - Focus on Metagenomics and Metabonomics

The human gut microbiota comprises a diverse microbial consortium closely co-evolved with the human genome and diet. The importance of the gut microbiota in regulating human health and disease has however been largely overlooked due to the inaccessibility of the intestinal habitat, the complexity of the gut microbiota itself and the fact that many of its members resist cultivation and are in fact new to science. However, with the emergence of 16S rRNA molecular tools and "post-genomics" high resolution technologies for examining microorganisms as they occur in nature without the need for prior laboratory culture, this limited view of the gut microbiota is rapidly changing. This review will discuss the application of molecular microbiological tools to study the human gut microbiota in a culture independent manner. Genomics or metagenomics approaches have a tremendous capability to generate compositional data and to measure the metabolic potential encoded by the combined genomes of the gut microbiota. Another post-genomics approach, metabonomics, has the capacity to measure the metabolic kinetic or flux of metabolites through an ecosystem at a particular point in time or over a time course. Metabonomics thus derives data on the function of the gut microbiota in situ and how it responds to different environmental stimuli e. g. substrates like prebiotics, antibiotics and other drugs and in response to disease. Recently these two culture independent, high resolution approaches have been combined into a single "transgenomic" approach which allows correlation of changes in metabolite profiles within human biofluids with microbiota compositional metagenomic data. Such approaches are providing novel insight into the composition, function and evolution of our gut microbiota.

The influence of age and gender on food choice: a focus group exploration

With 25% of the UK population predicted to be obese by 2010, the costs to individuals and society are set to rise. Due to the extra economic and social pressures obesity causes, there is an increasing need to understand what motivates and prevents consumers from eating a healthy diet so as to be able to tailor policy interventions to specific groups in society. In so doing, it is important to explore potential variations in attitudes, motivation and behaviour as a function of age and gender. Both demographic factors are easily distinguished within society and a future intervention study which targets either, or both, of these would likely be both feasible and cost-effective for policy makers. As part of a preliminary study, six focus groups (total n = 43) were conducted at the University of Reading in November 2006, with groups segmented on the basis of age and gender. In order to gather more sensitive information, participants were also asked to fill out a short anonymous questionnaire before each focus group began, relating to healthy eating, alcohol consumption and body dissatisfaction. Making use of thematic content analysis, results suggested that most participants were aware of the type of foods that contribute to a healthy diet and the importance of achieving a healthy balance within a diet. However, they believed that healthy eating messages were often conflicting, and were uncertain about where to find information on the topic. Participants believed that the family has an important role in educating children about eating habits. Despite these similarities, there were a number of key differences among the groups in terms of their reasons for making food choices. Older participants (60+ years old) were more likely to make food choices based on health considerations. Participants between the ages of 18–30 were less concerned with this link, and instead focused on issues of food preparation and knowledge, prices and time. Younger female participants said they had more energy when they ate healthier diets; however, very often their food choices related to concern with their appearance. Older female participants also expressed this concern within the questionnaire, rather than in the group discussions. Overall, these results suggest that consumer motivations for healthy eating are diverse and that this must be considered by government, retailers and food producers.

Automatic internal calibration in liquid chromatography/Fourier transform ion cyclotron resonance mass spectrometry of protein digests

Accurately measured peptide masses can be used for large-scale protein identification from bacterial whole-cell digests as an alternative to tandem mass spectrometry (MS/MS) provided mass measurement errors of a few parts-per-million (ppm) are obtained. Fourier transform ion cyclotron resonance (FTICR) mass spectrometry (MS) routinely achieves such mass accuracy either with internal calibration or by regulating the charge in the analyzer cell. We have developed a novel and automated method for internal calibration of liquid chromatography (LC)/FTICR data from whole-cell digests using peptides in the sample identified by concurrent MS/MS together with ambient polydimethyl-cyclosiloxanes as internal calibrants in the mass spectra. The method reduced mass measurement error from 4.3 +/- 3.7 ppm to 0.3 +/- 2.3 ppm in an E. coli LC/FTICR dataset of 1000 MS and MS/MS spectra and is applicable to all analyses of complex protein digests by FTICRMS. Copyright (c) 2006 John Wiley & Sons, Ltd.

Studying the Human Gut Microbiota in the Trans-Omics Era - Focus on Metagenomics and Metabonomics

The human gut microbiota comprises a diverse microbial consortium closely co-evolved with the human genome and diet. The importance of the gut microbiota in regulating human health and disease has however been largely overlooked due to the inaccessibility of the intestinal habitat, the complexity of the gut microbiota itself and the fact that many of its members resist cultivation and are in fact new to science. However, with the emergence of 16S rRNA molecular tools and "post-genomics" high resolution technologies for examining microorganisms as they occur in nature without the need for prior laboratory culture, this limited view of the gut microbiota is rapidly changing. This review will discuss the application of molecular microbiological tools to study the human gut microbiota in a culture independent manner. Genomics or metagenomics approaches have a tremendous capability to generate compositional data and to measure the metabolic potential encoded by the combined genomes of the gut microbiota. Another post-genomics approach, metabonomics, has the capacity to measure the metabolic kinetic or flux of metabolites through an ecosystem at a particular point in time or over a time course. Metabonomics thus derives data on the function of the gut microbiota in situ and how it responds to different environmental stimuli e.g. substrates like prebiotics, antibiotics and other drugs and in response to disease. Recently these two culture independent, high resolution approaches have been combined into a single "transgenomic" approach which allows correlation of changes in metabolite profiles within human biofluids with microbiota compositional metagenomic data. Such approaches are providing novel insight into the composition, function and evolution of our gut microbiota.

The automatic identification and prioritisation of criminal networks from police crime data

The identification of criminal networks is not a routine exploratory process within the current practice of the law enforcement authorities; rather it is triggered by specific evidence of criminal activity being investigated. A network is identified when a criminal comes to notice and any associates who could also be potentially implicated would need to be identified if only to be eliminated from the enquiries as suspects or witnesses as well as to prevent and/or detect crime. However, an identified network may not be the one causing most harm in a given area.. This paper identifies a methodology to identify all of the criminal networks that are present within a Law Enforcement Area, and, prioritises those that are causing most harm to the community. Each crime is allocated a score based on its crime type and how recently the crime was committed; the network score, which can be used as decision support to help prioritise it for law enforcement purposes, is the sum of the individual crime scores.

Semi-automatic annotation and retrieval of visual content using the topic map technology

There are still major challenges in the area of automatic indexing and retrieval of multimedia content data for very large multimedia content corpora. Current indexing and retrieval applications still use keywords to index multimedia content and those keywords usually do not provide any knowledge about the semantic content of the data. With the increasing amount of multimedia content, it is inefficient to continue with this approach. In this paper, we describe the project DREAM, which addresses such challenges by proposing a new framework for semi-automatic annotation and retrieval of multimedia based on the semantic content. The framework uses the Topic Map Technology, as a tool to model the knowledge automatically extracted from the multimedia content using an Automatic Labelling Engine. We describe how we acquire knowledge from the content and represent this knowledge using the support of NLP to automatically generate Topic Maps. The framework is described in the context of film post-production.