Ranking method for optimizing precision/recall of content-based image retrieval

The ranking method is a key element of Content-based Image Retrieval (CBIR) system, which can affect the final retrieval performance. In the literature, previous ranking methods based on either distance or probability do not explicitly relate to precision and recall, which are normally used to evaluate the performance of CBIR systems. In this paper, a novel ranking method based on relative density is proposed to improve the probability based approach by ranking images in the class. The proposed method can achieve optimal precision and recall. The experiments conducted on a large photographic collection show significant improvements of retrieval performance.

Measurement precision and evaluation of the diameter profiles of single wool fibers

A recent model of the Single Fiber Analyzer 3001 (SIFAN3001) was firstly employed to obtain the single wool fiber diameter profiles (SfFDPs) at multiple orientations. The results showed that using SIFAN3001 to measure fiber diameter at four orientations for 50 single fibers randomly sub-sampled from each mid-side sample can produce average fiber diameter profiles (AS fFDPs) of fibers within staples. Within the testing regime used, the precision estimates for the total samples were ±1.3 µm for the mean fiber diameter of staples and 1.4 µm for the average fiber diameter of the AS fFDPs at each scanned step in the diameter profile. The mean diameter ratio (ellipticity) obtained from the four orientations was 1.08±0.01, confirming that the Merino wool fibers under review were elliptical rather than circular. The elliptical morphology of wool fibers and the precision of the fiber diameter measurement at each point along a fiber will be considered in the development of a mechanical model of Staple Strength testing.

Checks and cheques : implementing a population health and recall system to improve coverage of patients with diabetes in a rural general practice

Identification of all diabetic patients in the population is essential if diabetic care is to be effective in achieving the targets of the St Vincent Declaration.¹ The challenge therefore is to establish population based monitoring and control systems by means of state of the art technology in order to achieve quality assurance in the provision of care for patients with diabetes. ^2,3 Disease management receives extensive international support as the most appropriate approach to organising and delivering healthcare for chronic conditions like diabetes.⁴ This approach is achieved through a combination of guidelines for practice, patient education, consultations and follow up using a planned team approach and a strong focus on continuous quality improvement using information technology. ^5,6 The current software (Medical Director) could not easily meet these requirements which led us to adopt a trial of Ferret. In designing this project we used change management⁷ and the plan, do, study, act cycle⁸ illustrated in Diagram 1.

The measurement precision and evaluation of the diameter profiles of single wool fibres

A recent model of Single Fibre Analyser 3001 (SIFAN3001) was firstly employed to obtain the single wool fibre diameter profiles (SfFDP’s) at multiple orientations. The results showed that using SIFAN3001 to measure fibre diameter at four orientations for 50 single fibres randomly sub-sampled from each mid-side sample can produce average fibre diameter profiles (ASfFDP’s) of fibres within staples. Within the testing regime used, the precision estimates for the total samples were ±1.3µm for mean fibre diameter of staples and ±1.4µm for average fibre diameter of the ASfFDP’s at each scanned step in the diameter profile. The mean diameter ratio (ellipticity) obtained from the four orientations was 1.08±0.01, confirming that the Merino wool fibres under review were elliptical rather than circular. The elliptical morphology of wool fibres and the precision of fibre diameter measurement at each point along a fibre will be considered in the development of a mechanical model of Staple Strength testing.

Application of rank correlation, clustering and classification in information security

This article is devoted to experimental investigation of a novel application of a clustering technique introduced by the authors recently in order to use robust and stable consensus functions in information security, where it is often necessary to process large data sets and monitor outcomes in real time, as it is required, for example, for intrusion detection. Here we concentrate on a particular case of application to profiling of phishing websites. First, we apply several independent clustering algorithms to a randomized sample of data to obtain independent initial clusterings. Silhouette index is used to determine the number of clusters. Second, rank correlation is used to select a subset of features for dimensionality reduction. We investigate the effectiveness of the Pearson Linear Correlation Coefficient, the Spearman Rank Correlation Coefficient and the Goodman--Kruskal Correlation Coefficient in this application. Third, we use a consensus function to combine independent initial clusterings into one consensus clustering. Fourth, we train fast supervised classification algorithms on the resulting consensus clustering in order to enable them to process the whole large data set as well as new data. The precision and recall of classifiers at the final stage of this scheme are critical for the effectiveness of the whole procedure. We investigated various combinations of several correlation coefficients, consensus functions, and a variety of supervised classification algorithms.

The effects of controlled language processing on listening comprehension and recall

This study seeks to determine the possible interactions between listening proficiency and the state of strategic self-awareness; second, and more importantly, to investigate the effects of learned strategies on listening comprehension and recall; and finally to describe the most common real-time listening comprehension problems faced by EFL learners and to compare the differences between learners with different listening abilities. After ten training sessions, an assessment was made to see whether or not well-learned strategies could provide students with ample opportunity to practice the comprehension and recall processes. The analyses of the data revealed the causes of ineffective low-level processing and provided insights to solve the problems of parsing. Moreover, the study reveals that explicit instruction of cognitive and metacognitive strategies is needed if a syllabus wishes to help learners improve their listening comprehension and become more-proficient at directing their own learning and development as L2 listeners.

The effect of product placement in computer games on brand attitude and recall

Media fragmentation and proliferation, in concert with declining television advertising efficacy, has engendered interest in developing more effective ways to reach consumers – particularly non-users of a brand. This study explores the effect of active product placement in computer games on both brand attitude (Abrand) and recall. Findings suggest that exposure to a particular brand in a computer game can increase Abrand among consumers whose pre-existing attitude towards the brand in question is fairly low. It was concluded that product placement within computer games is an effective means of fostering high spontaneous brand recall and even of influencing consumers less positively predisposed towards a brand (analogous to non-users). These findings have promising managerial implications for firms looking to grow their customer base through acquisition and conversion.

Spaced repetition tool for improving long-term memory retention and recall of collected personal experiences

A variety of electronic displays available for home use creates opportunities for intelligent applications. This paper presents a semi-passive photo reviewing tool for consolidating memories of experiences utilizing personal picture libraries. A form of spaced repetition algorithm is used to create visual journeys which link photos together around a user-chosen central theme. Systematically reviewing images from positive personal experiences can be useful to remember significant events, as well as to balance out stressful events in our lives. The design exploits existing digital home displays and aims to improve usage of media collections

Listerine – for the bridesmaid who’s never a bride: disparaging humour increases brand attitude and recall among the powerless

Purpose: While the potential benefits of integrating humour into advertisements are widely understood, the reasons why these effects emerge are not. Drawing on literature about the impact of psychological feelings of power, this research aims to examine how power motivation interacts with the presence of disparaging humour in ads to influence ad-related outcomes.

Design/methodology/approach: Following the measurement (Study 1) or manipulation (Study 2) of power motivation, participants viewed an ad featuring either disparaging humour or one of the following alternatives: no humour (Study 1) or non-disparaging humour (Study 2). Sense of superiority, brand attitude, ad claim recall and the perceived humorousness of the ad were then assessed.

Findings: Featuring disparaging humour in an ad increased participants’ sense of superiority, but only among those with high power motivation. Among such participants, this heightened sense of superiority increased the perceived humorousness of the disparaging humour (Studies 1 and 2), induced more favourable attitudes towards the brand featured in the ad (Studies 1 and 2) and enhanced ad claim recall (Study 2). These effects did not, however, extend to ads featuring non-disparaging humour (Study 2), indicating that it was the presence of disparaging humour, and not humour per se, that was responsible for these effects.

Originality/value: These findings break open the “black box” of humour by identifying why consumers perceive disparaging humorous content to be funny, when this effect will occur and what impact this will have on advertising-related outcomes.

Automatic parameters selection for eigenfaces

In this paper, we investigate the parameters selection for Eigenfaces. Our focus is on the eigenvectors and threshold selection issues. We will propose a systematic approach in selecting the eigenvectors based on relative errors of the eigenvalues for the covariance matrix. In addition, we have proposed a method for selecting the classification threshold that utilizes the information obtained from the training data set. Experimentation was conducted on two benchmark face databases, ORL and AMP, with results indicating that the proposed automatic eigenvectors and threshold selection methods produce better recognition performance in terms of precision and recall rates. Furthermore, we show that the eigenvector selection method outperforms energy and stretching dimension methods in terms of selected number of eigenvectors and computation cost.

Threshold dynamic time warping for spatial activity recognition

Non-invasive spatial activity recognition is a difficult task, complicated by variation in how the same activities are conducted and furthermore by noise introduced by video tracking procedures. In this paper we propose an algorithm based on dynamic time warping (DTW) as a viable method with which to quantify segmented spatial activity sequences from a video tracking system. DTW is a widely used technique for optimally aligning or warping temporal sequences through minimisation of the distance between their components. The proposed algorithm threshold DTW (TDTW) is capable of accurate spatial sequence distance quantification and is shown using a three class spatial data set to be more robust and accurate than DTW and the discrete hidden markov model (HMM). We also evaluate the application of a band dynamic programming (DP) constraint to TDTW in order to reduce extraneous warping between sequences and to reduce the computation complexity of the approach. Results show that application of a band DP constraint to TDTW improves runtime performance significantly, whilst still maintaining a high precision and recall.

Application of the DCT energy histogram for face recognition

In this paper, we investigate the face recognition problem via energy histogram of the DCT coefficients. Several issues related to the recognition performance are discussed, In particular the issue of histogram bin sizes and feature sets. In addition, we propose a technique for selecting the classification threshold incrementally. Experimentation was conducted on the Yale face database and results indicated that the threshold obtained via the proposed technique provides a balanced recognition in term of precision and recall. Furthermore, it demonstrated that the energy histogram algorithm outperformed the well-known Eigenface algorithm.

An application of novel clustering technique for information security

This article presents experimental results devoted to a new application of the novel clustering technique introduced by the authors recently. Our aim is to facilitate the application of robust and stable consensus functions in information security, where it is often necessary to process large data sets and monitor outcomes in real time, as it is required, for example, for intrusion detection. Here we concentrate on the particular case of application to profiling of phishing websites. First, we apply several independent clustering algorithms to a randomized sample of data to obtain independent initial clusterings. Silhouette index is used to determine the number of clusters. Second, we use a consensus function to combine these independent clusterings into one consensus clustering . Feature ranking is used to select a subset of features for the consensus function. Third, we train fast supervised classification algorithms on the resulting consensus clustering in order to enable them to process the whole large data set as well as new data. The precision and recall of classifiers at the final stage of this scheme are critical for effectiveness of the whole procedure. We investigated various combinations of three consensus functions, Cluster-Based Graph Formulation (CBGF), Hybrid Bipartite Graph Formulation (HBGF), and Instance-Based Graph Formulation (IBGF) and a variety of supervised classification algorithms. The best precision and recall have been obtained by the combination of the HBGF consensus function and the SMO classifier with the polynomial kernel.

Automatic parameter selection for Eigenfaces

In this paper, we investigate the parameter selection issues for Eigenfaces. Our focus is on the eigenvectors and threshold selection issues. We propose a systematic approach in selecting the eigenvectors based on the relative errors of the eigenvalues. In addition, we have designed a method for selecting the classification threshold that utilizes the information obtained from the training database effectively. Experimentation was conducted on the ORL and AMP face databases with results indicating that the automatic eigenvectors and threshold selection methods provide an optimum recognition in terms of precision and recall rates. Furthermore, we show that the eigenvector selection method outperforms energy and stretching dimension methods in terms of selected number of eigenvectors and computation cost.