An EM-based algorithm for clustering data streams in sliding windows

Cluster analysis has played a key role in data understanding. When such an important data mining task is extended to the context of data streams, it becomes more challenging since the data arrive at a mining system in one-pass manner. The problem is even more difficult when the clustering task is considered in a sliding window model which requiring the elimination of outdated data must be dealt with properly. We propose SWEM algorithm that exploits the Expectation Maximization technique to address these challenges. SWEM is not only able to process the stream in an incremental manner, but also capable to adapt to changes happened in the underlying stream distribution.

Incremental and adaptive clustering stream data over sliding window

Cluster analysis has played a key role in data stream understanding. The problem is difficult when the clustering task is considered in a sliding window model in which the requirement of outdated data elimination must be dealt with properly. We propose SWEM algorithm that is designed based on the Expectation Maximization technique to address these challenges. Equipped in SWEM is the capability to compute clusters incrementally using a small number of statistics summarized over the stream and the capability to adapt to the stream distribution’s changes. The feasibility of SWEM has been verified via a number of experiments and we show that it is superior than Clustream algorithm, for both synthetic and real datasets.

Enhancing Chinese SME performance through innovative HR practices

Purpose – The purpose of this paper is to show how understanding of human resource (HR) management practices which have been adopted in the emerging markets such as that in China is particularly interesting to academia and management practitioners. The purpose of this study is to shed some light on the implementation of innovative HR practices among 74 Chinese small and medium-sized enterprises (SMEs) and to explain how the HR practices influence their firm performance.

Design/methodology/approach – Cluster analysis is used to group Chinese SMEs according to their adoption of innovative human resource (HR) practices and examine how the practices are associated with HR outcomes and firm performance.

Findings – It is found that the membership of clusters is influenced by several factors, including ownership, age and size of firms. These characteristics have influenced the motivation, capacity and ability of firms in the sample to adopt high performance human resource practices. The extent to which firms have adopted innovative human resource practices is shown to be closely associated with human resource outcomes and firm performance.

Originality/value – The key implication is that SMEs, especially those rapidly developing domestic and collectively owned small firms, as well as those state-owned enterprises in China, may see clearly the benefits of devoting greater attention to HR practices to achieve their future growth potential.

Profiling contemporary marketing practices in Bangladesh

Purpose – The purpose of this paper is to explore how different types of firms relate to their markets in
terms of contemporary marketing practices (CMP) in an emerging country, Bangladesh. Additionally,
the paper also examines the various marketing performance measures used by Bangladeshi firms.
Design/methodology/approach – The CMP survey was used on 165 marketing managers chosen from
a range of industrial sectors as a basis for data collection. Data were analyzed using cluster analysis and other descriptive statistics.
Findings – The study found that a pluralistic marketing approach is predominant among the majority of
the Bangladeshi firms, while few other firms also practise transactional marketing. Results also reveal
that Bangladeshi firms apply a blend of performance indicators rather than relying on specific financialor
client-based measures to evaluate business success.
Research limitations/implications – The present study provides a benchmark for future studies on
CMP in emerging/developing countries and inspires further research designed to deepen
understanding about how marketing is practised in emerging markets and how they may differ from
developed markets.
Originality/value – Since very few studies have been conducted regarding CMP by incorporating both
business-to-business and consumer goods/services firms for a specific country into an emerging
market, this study adds a new dimension to the horizon of CMPs.

A validity-driven approach to the understanding of the personal and societal burden of low back pain : development of a conceptual and measurement model

Introduction: While the importance and magnitude of the burden of low back pain upon the individual is well recognized, a systematic understanding of the impact of the condition on individuals is currently hampered by the lack of an organized understanding of what aspects of a person’s life are affected and the lack of comprehensive measures for these effects. The aim of the present study was to develop a conceptual and measurement model of the overall burden of low back pain from the individual’s perspective using a validity-driven approach.
Methods: To define the breadth of low back pain burden we conducted three concept-mapping workshops to generate an item pool. Two face-to-face workshops (Australia) were conducted with people with low back pain and clinicians and policy-makers, respectively. A third workshop (USA) was held with international multidisciplinary experts. Multidimensional scaling, cluster analysis, participant input and thematic analyses organized participants’ ideas into clusters of ideas that then informed the conceptual model.
Results: One hundred and ninety-nine statements were generated. Considerable overlap was observed between groups, and four major clusters were observed - Psychosocial, Physical, Treatment and Employment - each with between two and six subclusters. Content analysis revealed that elements of the Psychosocial cluster were sufficiently distinct to be split into Psychological and Social, and a further cluster of elements termed Positive Effects also emerged. Finally, a hypothesized structure was proposed with six domains and 16 subdomains. New domains not previously considered in the back pain field emerged for psychometric verification: loss of independence, worry about the future, and negative or discriminatory actions by others.
Conclusions: Using a grounded approach, an explicit a priori and testable model of the overall burden of low back pain has been proposed that captures the full breadth of the burden experienced by patients and observed by experts.

Hierarchical summarization techniques for network traffic

In today’s high speed networks it is becoming increasingly challenging for network managers to understand the nature of the traffic that is carried in their network. A major problem for traffic analysis in this context is how to extract a concise yet accurate summary of the relevant aggregate traffic flows that are present in network traces. In this paper, we present two summarization techniques to minimize the size of the traffic flow report that is generated by a hierarchical cluster analysis tool. By analyzing the accuracy and compaction gain of our approach on a standard benchmark dataset, we demonstrate that our approach achieves more accurate summaries than those of an existing tool that is based on frequent itemset mining.

Does self monitoring of blood blucose as opposed to urinalysis provide additional benefit in patients newly diagnosed with type 2 diabetes receiving structured education? The DESMOND SMBG randomised controlled trial protocol

Background
The benefit of self-monitoring of blood glucose (SMBG) in people with type 2 diabetes on diet or oral agents other than sulphonylureas remains uncertain. Trials of interventions incorporating education about self-monitoring of blood glucose have reported mixed results. A recent systematic review concluded that SMBG was not cost-effective. However, what was unclear was whether a cheaper method of self-monitoring (such as urine glucose monitoring) could produce comparable benefit and patient acceptability for less cost.

Methods/Design
The DESMOND SMBG trial is comparing two monitoring strategies (blood glucose monitoring and urine testing) over 18 months when incorporated into a comprehensive self-management structured education programme. It is a multi-site cluster randomised controlled trial, conducted across 8 sites (7 primary care trusts) in England, UK involving individuals with newly diagnosed Type 2 diabetes.

The trial has 80% power to demonstrate equivalence in mean HbA1c (the primary end-point) at 18 months of within ± 0.5% assuming 20% drop out and 20% non-consent. Secondary end-points include blood pressure, lipids, body weight and psychosocial measures as well as a qualitative sub-study.

Practices were randomised to one of two arms: participants attend a DESMOND programme incorporating a module on self-monitoring of either urine or blood glucose. The programme is delivered by accredited educators who received specific training about equipoise. Biomedical data are collected and psychosocial scales completed at baseline, and 6, 12, and 18 months post programme. Qualitative research with participants and educators will explore views and experiences of the trial and preferences for methods of monitoring.

Discussion
The DESMOND SMBG trial is designed to provide evidence to inform the debate about the value of self-monitoring of blood glucose in people with newly diagnosed type 2 diabetes. Strengths include a setting in primary care, a cluster design, a health economic analysis, a comparison of different methods of monitoring while controlling for other components of training within the context of a quality assured structured education programme and a qualitative sub-study.

The Obsessive Beliefs Questionnaire (OBQ) : Examination in nonclinical samples and development of a short version

The study aimed to examine the factor structure of the Obsessive Beliefs Questionnaire (OBQ), the most widely used measure of dysfunctional beliefs in obsessive—compulsive disorder (OCD). Multiple exploratory methods (exploratory factor analysis, cluster analysis by variable, multidimensional scaling) were used to examine the questionnaire. Confirmatory factor analyses were also performed in two large nonclinical samples from Australia (N = 1,234) and Israel ( N = 617). Our analyses suggested a four-factor solution with 38 items, where threat and responsibility formed separate dimensions (the “OBQ-TRIP”). This version had superior fit statistics across the two divergent confirmatory samples, when compared with four alternative models suggested by previous authors. Of the OBQ dimensions, the threat scale correlated most strongly with OCD symptom measures, even when controlling for depression. A short, 20-item version of the scale is offered for further study. Implications and limitations are discussed.

Conservation of genetic diversity in British populations of the diploid endemic Coincya monensis ssp monensis (Isle of Man Cabbage): the risk of hybridisation with the tetraploid alien, Coincya monensis ssp cheiranthos

Coincya monensis is represented in the British flora by two, cytologically distinct subspecies. Coincya monensis ssp monensis is an endemic diploid with a coastal sand dune distribution that includes a number of isolated populations. Coincya monensis ssp cheiranthos is a tetraploid alien, well established in South Wales in early successional habitats. Both subspecies share similar life form traits, flowering times and pollinators. Cluster analysis and phylogenetic reconstruction based on sequences of the mitochondrial nad4 gene confirmed the distinction between alien and endemic taxa. Tetraploid populations carry more polymorphic RAPDs loci and their genetic diversity is partitioned more within than among populations. In contrast, C. monensis ssp monensis has a distinct population genetic structure. Analysis of the multilocus genetic data confirmed a structure of genetically isolated, endemic population clusters in Scotland, Arran, the Isle of Man and South Wales. Experimental hybridisation showed the two subspecies are interfertile. Multivariate analysis of RAPDs data resolved hybrids between alien and endemic clusters and hybrids contained a proportion of alien-specific polymorphic loci. Hybrids of alien maternal parentage contained the mitochondrial nad4 sequence characteristic of the alien subspecies. Since the alien subspecies can invade mobile sand dune communities from urban sites and compete for pollinators, there is a risk that alien and endemic populations will mix and introgress. Conservation of endemic genetic diversity in Britain will require protection for all C. monensis ssp monensis populations. Currently, the most disjunct endemic population in South Wales is most at risk from introgression.

Discovering consumer personality clusters in prestige sensitivity and fashion consciousness context

This article aims to discover the differences among consumer personality clusters in regard to their extent of fashion consciousness and prestige sensitivity. Data were collected from 251 undergraduate students using self-administered questionnaires. Cluster analysis and MANOVA were employed to assess whether significant differences exist among four personality clusters. The study used the Big Five scale items to measure consumer personality and found that respondents who score low on the “openness to experience” dimension tend to be less prestige sensitive, and those who score high on “extraversion,” “agreeableness,” and “conscientiousness” tend to be highly fashion conscious.

The role of external influences in high involvement purchase behaviour

Purpose
The purpose of this paper is to discover the consumer decision-making style clusters within the context of automobile purchases in Australia. It also examines the differences between consumer decision-making styles in terms of the importance given to external influences, such as importance of dealers, importance of friends/family members, number of cars test driven, time spent researching final decision and importance of information sources (e.g. internet, magazines, TV ads, word of mouth, etc.), prior to making their final purchase decision.

Design/methodology/approach
Data were collected from 209 respondents using self-administered questionnaires. Cluster analysis and ANOVA were employed to identify and analyse the differences between consumer decision-making style clusters. Consumer Styles Inventory (CSI; Sproles and Kendall, 1986) was used to measure respondents’ consumer decision-making styles in relation to automobile purchases.

Findings
Three clusters were identified from the analysis, namely “innovative informed”, “rational confused”, and “traditional habitual”. Significant differences were found between the clusters in terms of the average time they spent with each car dealer, the time they spent on researching final decision and the importance of consulting with family members prior to making their final purchase decision.

Practical implications
The paper found that some consumers rely heavily on friends/families and dealers as the most important sources of information. Other sources of information consumers use include television advertisements, newspapers, billboards and magazines. Based on the findings, marketers should focus on providing similar types of information/messages by using these above-mentioned sources when communicating with this type of consumers. Dealers could be trained to spend time explaining product features and benefits in full with these consumers and their friends and family members whom they are likely to bring along before making the final purchase decision.

Originality/value
The findings of this study have extended the knowledge by determining the impact of external influences on consumer decision-making styles using the CSI in context of specific product which is yet to be known in relation to Australian automobile consumers.

In-home monitoring of older adults with vision impairment: exploring patients', caregivers' and professionals' views

Objective
To develop a conceptual framework for the design of an in-home monitoring system (IMS) based on the requirements of older adults with vision impairment (VI), informal caregivers and eye-care rehabilitation professionals.

Materials and Methods
Concept mapping, a mixed-methods statistical research tool, was used in the construction of the framework. Overall, 40 participants brainstormed or sorted and rated 83 statements concerning an IMS for older adults with VI. Multidimensional scaling and hierarchical cluster analysis were employed to construct the framework. A questionnaire yielded further insights into the views of a wider sample of older adults with VI (n=78) and caregivers (n=25) regarding IMS.

Results
Concept mapping revealed a nine-cluster model of IMS-related aspects including affordability, awareness of system capabilities, simplicity of installation, operation and maintenance, system integrity and reliability, fall detection and safe movement, user customization, user preferences regarding information delivery, and safety alerts for patients and caregivers. From the questionnaire, independence, safety and fall detection were the most commonly reported reasons for older adults and caregivers to accept an IMS. Concerns included cost, privacy, security of the information obtained through monitoring, system accuracy, and ease of use.

Discussion
Older adults with VI, caregivers and professionals are receptive to in-home monitoring, mainly for fall detection and safety monitoring, but have concerns that must be addressed when developing an IMS.

Conclusion
Our study provides a novel conceptual framework for the design of an IMS that will be maximally acceptable and beneficial to our ageing and vision-impaired population.

Development of a whole agency approach to market segmentation in parks

Park agencies must plan to accommodate a diversity of visitors in order to satisfy visitor expectations and encourage future visitation. This study applies a market segmentation approach to develop a visitor typology that is effective across a broad spectrum of parks and applicable to a range of priorities, both strategic and operational, within park management agencies. Over a four-year period, data was sourced from over 11,000 interviews conducted at 33 diverse Australian national and metropolitan parks managed by the agency Parks Victoria. Factor analysis and cluster analysis was used to identify seven distinct visitor segments on the basis of numerous variables including, crucially, benefits sought. The applied and theoretical contributions of this study to the parks literature are discussed.

‘Subtypes’ in the presentation of autistic traits in the general adult population

The present study examined the presentation of autistic traits in a large adult population sample (n = 2,343). Cluster analysis indicated two subgroups with clearly distinguishable trait profiles. One group (n = 1,059) reported greater social difficulties and lower detail orientation, while the second group (n = 1,284) reported lesser social difficulties and greater detail orientation. We also report a three-factor solution for the autism-spectrum quotient, with two, related, social-themed factors (Sociability and Mentalising) and a third non-social factor that varied independently (Detail Orientation). These results indicate that different profiles of autistic characteristics tend to occur in the adult nonclinical population. Research into nonclinical variance in autistic features may benefit by considering social- and detail-related trait domains independently.

Segmenting Australian online panellists based on volunteering motivations

Purpose– The purpose of this paper is to seek to assess whether online commercial panel volunteering can be segmented based on their motivations, using the volunteer functions inventor. The authors also investigate whether segments exist which differ in demographic characteristics.Design/methodology/approach– The authors survey 484 Australian online panel volunteers using a adapted version of the 30 item of the volunteer function inventory (VFI) scale developed by Clary et al. (1998). Data were analysed using confirmatory factor analysis (CFA) and cluster analysis, as well as ANOVA and χ2 test comparisons of demographics between clusters.Findings– CFA verifies that the VFI scale is suitable instrument to gauge online participants’ motivations. Cluster analysis produced a five-cluster solution, where respondents with low motivations overall comprised the largest grouping. Segments are interpreted by assessing the difference between the total sample average and the segment profile. The examination also identifies that the only demographic factor that varies across the five clusters is “respondents” employment status”.Research limitations/implications– Future research could explore if differences in segments result in differences in online participation. The high number or respondents with low motivations may explain the relatively high levels of churn that take place within online panels and as a result panel operators would need to continually attract new members. Further research could also investigate whether the levels of motivation change over time and if so what effect such variation would produce on respondents’ retention.Originality/value– Research on online panel respondents’ motivation is still limited and investigating online panellists’ motivation as volunteers is very important as it unveils, as in the study herein reported, that alternative types of respondents may be driven by different factors when joining an online panel (or completing a given survey). Recruitment strategies could, therefore, be shaped to suit the motivation of the different segments. By refining the matching between volunteers’ profiles and their motivation, managers could improve how volunteers are recruited, managed and retained.