Preference mining techniques for customer behavior analysis

The thesis has studied a number of critical problems in data mining for customer behavior analysis and has proposed novel techniques for better modeling of the customers’ decision making process, more efficient analysis of their travel behavior, and more effective identification of their emerging preference.

Web search activity data accurately predict population chronic disease risk in the USA

BACKGROUND: The WHO framework for non-communicable disease (NCD) describes risks and outcomes comprising the majority of the global burden of disease. These factors are complex and interact at biological, behavioural, environmental and policy levels presenting challenges for population monitoring and intervention evaluation. This paper explores the utility of machine learning methods applied to population-level web search activity behaviour as a proxy for chronic disease risk factors. METHODS: Web activity output for each element of the WHO's Causes of NCD framework was used as a basis for identifying relevant web search activity from 2004 to 2013 for the USA. Multiple linear regression models with regularisation were used to generate predictive algorithms, mapping web search activity to Centers for Disease Control and Prevention (CDC) measured risk factor/disease prevalence. Predictions for subsequent target years not included in the model derivation were tested against CDC data from population surveys using Pearson correlation and Spearman's r. RESULTS: For 2011 and 2012, predicted prevalence was very strongly correlated with measured risk data ranging from fruits and vegetables consumed (r=0.81; 95% CI 0.68 to 0.89) to alcohol consumption (r=0.96; 95% CI 0.93 to 0.98). Mean difference between predicted and measured differences by State ranged from 0.03 to 2.16. Spearman's r for state-wise predicted versus measured prevalence varied from 0.82 to 0.93. CONCLUSIONS: The high predictive validity of web search activity for NCD risk has potential to provide real-time information on population risk during policy implementation and other population-level NCD prevention efforts.

A pilot, multisite, randomized controlled trial of a self-directed coping skills training intervention for couples facing prostate cancer : accrual, retention, and data collection issues

PURPOSE: To examine the acceptability of the methods used to evaluate Coping-Together, one of the first self-directed coping skill intervention for couples facing cancer, and to collect preliminary efficacy data. METHODS: Forty-two couples, randomized to a minimal ethical care (MEC) condition or to Coping-Together, completed a survey at baseline and 2 months after, a cost diary, and a process evaluation phone interview. RESULTS: One hundred seventy patients were referred to the study. However, 57 couples did not meet all eligibility criteria, and 51 refused study participation. On average, two to three couples were randomized per month, and on average it took 26 days to enrol a couple in the study. Two couples withdrew from MEC, none from Coping-Together. Only 44 % of the cost diaries were completed, and 55 % of patients and 60 % of partners found the surveys too long, and this despite the follow-up survey being five pages shorter than the baseline one. Trends in favor of Coping-Together were noted for both patients and their partners. CONCLUSIONS: This study identified the challenges of conducting dyadic research, and a number of suggestions were put forward for future studies, including to question whether distress screening was necessary and what kind of control group might be more appropriate in future studies.

The impact of data realities on conservation planning

Conservation planning decisions are typically made on the basis of species distribution or occurrence data, which ideally would have complete spatial and taxonomic coverage. Agencies are constrained in the data they can collect, often pragmatically prioritising certain groups such as threatened species, or methods, such as volunteer surveys. This mismatch between goals and realities inevitably leads to bias and uncertainty in conservation planning outputs, yet few studies have assessed how data realities affect planning outputs. We conducted a sensitivity analysis on the Protection Index, a method for assessing conservation progress and priorities, using an extensive dataset on species occurrences and distributions derived from the Florida Natural Areas Inventory. Analyses revealed a high proportion of occurrence records for threatened species and certain taxonomic groups, reflecting the agencies' priorities. We performed a sensitivity analysis on conservation planning outputs, simulating a 'data poor' scenario typical of many real situations; we deleted increasing amounts of data in both a biased (exaggerating patterns observed) and unbiased (random) manner. We assessed the effects of data paucity and bias on the value of potential conservation sites, and planning priorities. Certain high value sites with only a few important species occurrences were more sensitive to data depletion than those with many occurrences. Data bias based on taxonomic bias was more influential to site value than threat rank. To maximise benefit from surveys from a planning perspective, it would be better to focus on poorly surveyed areas rather than adding occurrences in already well represented sites. This study demonstrates the importance of sensitivity analysis in conservation planning, and that the effects of uncertainty and data quality on planning decisions should not be ignored. © 2011 Elsevier Ltd.

Australian children's independent mobility levels: secondary analyses of cross-sectional data between 1991 and 2012

This study investigated changes in Australian children's independent mobility levels between1991 and 2012. Data from five cross-sectional studies conducted in 1991, 1993, 2010, 2011 and 2012 were analysed. Parent and child surveys were used to assess parental licences for independent mobility and actual independent mobility behaviour in children aged 8–13 years. Findings show declines in the proportion of young children (≤10 years of age) being allowed to travel home from school alone (1991: 68%, 1993: 50%, 2010: 43%, 2011: 45%, 2012: 31%) and travel on buses alone (1991: 31%, 1993: 15%, 2010: 8%, 2011: 6%, 2012: 9%). Furthermore, the proportion of children travelling independently to school decreased (1991: 61%, 1993: 42%, 2010: 31%, 2011: 32%, 2012: 32%). Significantly fewer girls than boys travelled independently to school at each time point (p ≤ .001). Overall, the findings suggest that Australian children's independent mobility levels declined between 1991 and 2012.

Age-specific changes in BMI and BMI distribution among Australian adults using cross-sectional surveys from 1980 to 2008

Background: Research efforts have focused mainly on trends in obesity among populations, or changes in mean body mass index (BMI), without consideration of changes in BMI across the BMI spectrum. Examination of age-specific changes in BMI distribution may reveal patterns that are relevant to targeting of interventions.

Methods: Using a synthetic cohort approach (which matches members of cross-sectional surveys by birth year) we estimated population representative annual BMI change across two time periods (1980 to 1989 and 1995 to 2008) by age, sex, socioeconomic position and quantiles of BMI. Our study population was a total of 27 349 participants from four nationally representative Australian health surveys; Risk Factor Prevalence Study surveys (1980 and 1989), the 1995 National Nutrition Survey and the 2007/8 National Health Survey.

Results: We found greater mean BMI increases in younger people, in those already overweight and in those with lower education. For men, age-specific mean annual BMI change was very similar in the 1980s and the early 2000s (P=0.39), but there was a recent slowing down of annual BMI gain for older women in the 2000s compared with their same-age counterparts in the 1980s (P<0.05). BMI change was not uniform across the BMI distribution, with different patterns by age and sex in different periods. Young adults had much greater BMI gain at higher BMI quantiles, thus adding to the increased right skew in BMI, whereas BMI gain for older populations was more even across the BMI distribution.

Conclusions: The synthetic cohort technique provided useful information from serial cross-sectional survey data. The quantification of annual BMI change has contributed to an understanding of the epidemiology of obesity progression and identified key target groups for policy attention—young adults, those who are already overweight and those of lower socioeconomic status.

A suite of visual languages for model-driven development of statistical surveys and services

Objective: To provide statistician end users with a visual language environment for complex statistical survey design and implementation. Methods: We have developed, in conjunction with professional statisticians, the Statistical Design Language (SDL), an integrated suite of visual languages aimed at supporting the process of designing statistical surveys, and its support environment, SDLTool. SDL comprises five diagrammatic notations: survey diagrams, data diagrams, technique diagrams, task diagrams and process diagrams. SDLTool provides an integrated environment supporting design, coordination, execution, sharing and publication of complex statistical survey techniques as web services. SDLTool allows association of model components with survey artefacts, including data sets, metadata, and statistical package analysis scripts, with the ability to execute elements of the survey design model to implement survey analysis. Results: We describe three evaluations of SDL and SDLTool: use of the notation by expert statistician to design and execute surveys; useability evaluation of the environment; and assessment of several generated statistical analysis web services. Conclusion: We have shown the effectiveness of SDLTool for supporting statistical survey design and implementation. Practice implications: We have developed a more effective approach to supporting statisticians in their survey design work.

A survey of recommendation techniques based on offline data processing

Recommendations based on offline data processing has attracted increasing attention from both research communities and IT industries. The recommendation techniques could be used to explore huge volumes of data, identify the items that users probably like, translate the research results into real-world applications and so on. This paper surveys the recent progress in the research of recommendations based on offline data processing, with emphasis on new techniques (such as temporal recommendation, graph-based recommendation and trust-based recommendation), new features (such as serendipitous recommendation) and new research issues (such as tag recommendation and group recommendation). We also provide an extensive review of evaluation measurements, benchmark data sets and available open source tools. Finally, we outline some existing challenges for future research.

Examining customer evaluations across different self-service technologies

Purpose – This study aims to examine the influence of different self-service technologies (SSTs) on customer satisfaction with and continued usage of SSTs. Specifically, it compares an interactive voice response (IVR) SST and an online SST from the same provider to assess how to manage these parallel SSTs.

Design/methodology/approach – A tracking study was used, beginning with a survey of n = 957 SST users to test a model pertaining to SST satisfaction across IVR and online SSTs. These SST users were then tracked over 12 months. The association between customer satisfaction with and continued usage of the SSTs was examined using behavioural data from the service provider.

Findings – While the overall model was found to be valid across both types of SSTs, perceptions of factors including ease of use, perceived control and reliability differed for IVR and online SSTs. Satisfaction with SSTs is linked with users’ continued use of SSTs, but is not a barrier to users’ adoption of newer SST forms.

Research limitations/implications – Highlighting the rapid developments in this field, a new SST was introduced by the provider to respondents during the 12-month tracking period, thus complicating the results. Further studies could include the customer purpose for using SSTs as a variable.

Practical implications – The findings offer support for organisations offering a suite of SSTs, even if they serve the same purpose. Customers evaluate SST types differently, and even satisfied SST users switch to different SSTs when they become available. Allowing customers to choose the SST that best suits them appears to be good practice.

Originality/value – This study develops a comprehensive model of customer SST satisfaction that is used to undertake a comparison of two different types of SSTs, which has been missing from prior research.

Accounting for uncertainty in volumes of seabed change measured with repeat multibeam sonar surveys

Seafloors of unconsolidated sediment are highly dynamic features; eroding or accumulating under the action of tides, waves and currents. Assessing which areas of the seafloor experienced change and measuring the corresponding volumes involved provide insights into these important active sedimentation processes. Computing the difference between Digital Elevation Models (DEMs) obtained from repeat Multibeam Echosounders (MBES) surveys has become a common technique to identify these areas, but the uncertainty in these datasets considerably affects the estimation of the volumes displaced. The two main techniques used to take into account uncertainty in volume estimations are the limitation of calculations to areas experiencing a change in depth beyond a chosen threshold, and the computation of volumetric confidence intervals. However, these techniques are still in their infancy and, as a result, are often crude, seldom used or poorly understood. In this article, we explored a number of possible methodological advances to address this issue, including: (1) using the uncertainty information provided by the MBES data processing algorithm CUBE, (2) adapting fluvial geomorphology techniques for volume calculations using spatially variable thresholds and (3) volumetric histograms. The nearshore seabed off Warrnambool harbour - located in the highly energetic southwest Victorian coast, Australia - was used as a test site. Four consecutive MBES surveys were carried out over a four-months period. The difference between consecutive DEMs revealed an area near the beach experiencing large sediment transfers - mostly erosion - and an area of reef experiencing increasing deposition from the advance of a nearby sediment sheet. The volumes of sediment displaced in these two areas were calculated using the techniques described above, both traditionally and using the suggested improvements. We compared the results and discussed the applicability of the new methodological improvements. We found that the spatially variable uncertainty derived from the CUBE algorithm provided the best results (i.e. smaller confidence intervals), but that similar results can be obtained using as a fixed uncertainty value derived from a reference area under a number of operational conditions.

Desenvolvimento e implantação de Data Base Marketing para apoiar estratégia de gestão de fidelização de clientes em empresa administradora de consórcios

o ambiente econômico atual tem exigido empenho das empresas em conhecer, interagir, diferenciar e personalizar cada vez mais produtos e serviços para os clientes. Este cenário requer ferramentas e modelos de gestão para gerenciar as relações com os clientes, com o objetivo de permitir que a empresa consiga perceber e responder rapidamente a exigências dos consumidores. Este trabalho revisa conceitos de CRM (Customer Relationschip Management ou Gerenciamento das Relações com os Clientes) e descreve a implementação de ferramenta de gestão de relacionamento com clientes em empresa de consórcio. O desenvolvimento do trabalho reflete uma necessidade apontada no planejamento estratégico da empresa, sendo que ferramentas de tecnologia de informação e software de banco de dados foram usadas como suporte aos propósitos da gestão empresarial. Como resultado do trabalho, a empresa está hoje atuando com um sistema de Data Base Marketing, o qual foi criado para auxiliar os profissionais envolvidos no processo de atendimento e gestão de relacionamento com clientes. O Data Base Marketing esta sendo utilizado para coletar dados de atendimento a clientes, tais como históricos de atendimento, dados cadastrais, perfil demográfico, perfil psicográfico e categoria de valor dos clientes. Durante o processo de interação com clientes, o sistema facilita o trabalho dos especialistas e permite melhorar a qualidade do atendimento aos clientes, contemplando necessidades dos diversos especialistas da empresa em assuntos como vendas, qualidade em serviços, finanças e gestão empresarial.O processo começou pela constituição de um grupo de trabalho interno para discutir estratégias e cronograma de implantação. A primeira decisão do grupo foi pelo desenvolvimento interno do software visando atender plenamente o "core business" da empresa. O processo começou pela constituição de um grupo de trabalho interno para discutir estratégias e cronograma de implantação. A primeira decisão do grupo foi pelo desenvolvimento interno do software visando atender plenamente o "core business" da empresa. O projeto contou com o conhecimento do negócio dos profissionais da empresa e auxilio de especialistas e consultores externos. O detalhamento do projeto, bem como os passos da pesquisa-ação, está descrito no corpo da dissertação.

A panel data approach to economic forecasting: the bias-corrected average forecast

In this paper, we propose a novel approach to econometric forecasting of stationary and ergodic time series within a panel-data framework. Our key element is to employ the bias-corrected average forecast. Using panel-data sequential asymptotics we show that it is potentially superior to other techniques in several contexts. In particular it delivers a zero-limiting mean-squared error if the number of forecasts and the number of post-sample time periods is sufficiently large. We also develop a zero-mean test for the average bias. Monte-Carlo simulations are conducted to evaluate the performance of this new technique in finite samples. An empirical exercise, based upon data from well known surveys is also presented. Overall, these results show promise for the bias-corrected average forecast.

Regulation and wage adjustment patterns : non parametric evidence from longitudinal data

This paper gives a first step toward a methodology to quantify the influences of regulation on short-run earnings dynamics. It also provides evidence on the patterns of wage adjustment adopted during the recent high inflationary experience in Brazil.The large variety of official wage indexation rules adopted in Brazil during the recent years combined with the availability of monthly surveys on labor markets makes the Brazilian case a good laboratory to test how regulation affects earnings dynamics. In particular, the combination of large sample sizes with the possibility of following the same worker through short periods of time allows to estimate the cross-sectional distribution of longitudinal statistics based on observed earnings (e.g., monthly and annual rates of change).The empirical strategy adopted here is to compare the distributions of longitudinal statistics extracted from actual earnings data with simulations generated from minimum adjustment requirements imposed by the Brazilian Wage Law. The analysis provides statistics on how binding were wage regulation schemes. The visual analysis of the distribution of wage adjustments proves useful to highlight stylized facts that may guide future empirical work.

Do tratamento das reclamações à gestão da insatisfação: o comportamento de empresas hoteleiras mediante a insatisfação de seus clientes

Consumer dissatisfaction, when properly handled, is a significant information source for the manager. Studies in this area allow broadening the understanding of certain customer attitudes and behaviors, such as loyalty, repurchase intention or satisfaction and trust increase. Above and beyond supporting consumer feedback, dissatisfaction can provide significant opportunities for organizational learning. Starting from dissatisfied customer information, companies can detect service flaws and develop new products. This work presents the results of an investigation on the behavior of businesses belonging to the hotel sector in Natal, RN, through the dissatisfaction of their customers. We have sought to map the main problems presented by customers to hotels, in the perception of managers and employees, as well as to understand both the process of dissatisfactionrelated data collection, analysis, and processing, and the utilization of such information by businesses. Beyond this, we have compared the habits of organizations to the company reaction approaches described in the literature: Complaint Handling, Complaint Management, and Dissatisfaction Management. The used methodology has been based on case study. Data was collected via indepth interviews with managers and employees in six hotels, two independent ones and four belonging to national and international hotel networks. We have also made use of documents provided by the organizations, such as guest complaint registers and reports from satisfaction surveys on which content analysis was subsequently performed. The results of the investigation point to a high level of awareness in the companies concerning the importance of consumer dissatisfaction. Even though the maximum grade in the procedure scale is not achieved, it has been observed that answer to dissatisfaction is given in planned and systematic form, geared towards consumer satisfaction and improvement of products and processes. Hotel businesses still have to look into other possibilities for mapping consumer dissatisfaction, which implies, among other aspects, articulation with a range of public and private organizations in such a way as to guarantee sustainability of touristic activities in the long term

Estudo sistemático de estrelas variáveis na era dos grandes surveys

The interest in the systematic analysis of astronomical time series data, as well as development in astronomical instrumentation and automation over the past two decades has given rise to several questions of how to analyze and synthesize the growing amount of data. These data have led to many discoveries in the areas of modern astronomy asteroseismology, exoplanets and stellar evolution. However, treatment methods and data analysis have failed to follow the development of the instruments themselves, although much effort has been done. In present thesis, we propose new methods of data analysis and two catalogs of the variable stars that allowed the study of rotational modulation and stellar variability. Were analyzed the photometric databases fromtwo distinctmissions: CoRoT (Convection Rotation and planetary Transits) and WFCAM (Wide Field Camera). Furthermore the present work describes several methods for the analysis of photometric data besides propose and refine selection techniques of data using indices of variability. Preliminary results show that variability indices have an efficiency greater than the indices most often used in the literature. An efficient selection of variable stars is essential to improve the efficiency of all subsequent steps. Fromthese analyses were obtained two catalogs; first, fromtheWFCAMdatabase we achieve a catalog with 319 variable stars observed in the photometric bands Y ZJHK. These stars show periods ranging between ∼ 0, 2 to ∼ 560 days whose the variability signatures present RR-Lyrae, Cepheids , LPVs, cataclysmic variables, among many others. Second, from the CoRoT database we selected 4, 206 stars with typical signatures of rotationalmodulation, using a supervised process. These stars show periods ranging between ∼ 0, 33 to ∼ 92 days, amplitude variability between ∼ 0, 001 to ∼ 0, 5 mag, color index (J - H) between ∼ 0, 0 to ∼ 1, 4 mag and spectral type CoRoT FGKM. The WFCAM variable stars catalog is being used to compose a database of light curves to be used as template in an automatic classifier for variable stars observed by the project VVV (Visible and Infrared Survey Telescope for Astronomy) moreover it are a fundamental start point to study different scientific cases. For example, a set of 12 young stars who are in a star formation region and the study of RR Lyrae-whose properties are not well established in the infrared. Based on CoRoT results we were able to show, for the first time, the rotational modulation evolution for an wide homogeneous sample of field stars. The results are inagreement with those expected by the stellar evolution theory. Furthermore, we identified 4 solar-type stars ( with color indices, spectral type, luminosity class and rotation period close to the Sun) besides 400 M-giant stars that we have a special interest to forthcoming studies. From the solar-type stars we can describe the future and past of the Sun while properties of M-stars are not well known. Our results allow concluded that there is a high dependence of the color-period diagram with the reddening in which increase the uncertainties of the age-period realized by previous works using CoRoT data. This thesis provides a large data-set for different scientific works, such as; magnetic activity, cataclysmic variables, brown dwarfs, RR-Lyrae, solar analogous, giant stars, among others. For instance, these data will allow us to study the relationship of magnetic activitywith stellar evolution. Besides these aspects, this thesis presents an improved classification for a significant number of stars in the CoRoT database and introduces a new set of tools that can be used to improve the entire process of the photometric databases analysis