Discovering user access pattern based on probabilistic latent factor model

There has been an increased demand for characterizing user access patterns using web mining techniques since the informative knowledge extracted from web server log files can not only offer benefits for web site structure improvement but also for better understanding of user navigational behavior. In this paper, we present a web usage mining method, which utilize web user usage and page linkage information to capture user access pattern based on Probabilistic Latent Semantic Analysis (PLSA) model. A specific probabilistic model analysis algorithm, EM algorithm, is applied to the integrated usage data to infer the latent semantic factors as well as generate user session clusters for revealing user access patterns. Experiments have been conducted on real world data set to validate the effectiveness of the proposed approach. The results have shown that the presented method is capable of characterizing the latent semantic factors and generating user profile in terms of weighted page vectors, which may reflect the common access interest exhibited by users among same session cluster.

A web recommendation technique based on probabilistic latent semantic analysis

Web transaction data between Web visitors and Web functionalities usually convey user task-oriented behavior pattern. Mining such type of click-stream data will lead to capture usage pattern information. Nowadays Web usage mining technique has become one of most widely used methods for Web recommendation, which customizes Web content to user-preferred style. Traditional techniques of Web usage mining, such as Web user session or Web page clustering, association rule and frequent navigational path mining can only discover usage pattern explicitly. They, however, cannot reveal the underlying navigational activities and identify the latent relationships that are associated with the patterns among Web users as well as Web pages. In this work, we propose a Web recommendation framework incorporating Web usage mining technique based on Probabilistic Latent Semantic Analysis (PLSA) model. The main advantages of this method are, not only to discover usage-based access pattern, but also to reveal the underlying latent factor as well. With the discovered user access pattern, we then present user more interested content via collaborative recommendation. To validate the effectiveness of proposed approach, we conduct experiments on real world datasets and make comparisons with some existing traditional techniques. The preliminary experimental results demonstrate the usability of the proposed approach.

Gray's model of personality and aggregate level factor analysis

Previous research shows that correlations tend to increase in magnitude when individuals are aggregated across groups. This suggests that uncorrelated constellations of personality variables (such as the primary scales of Extraversion and Neuroticism) may display much higher correlations in aggregate factor analysis. We hypothesize and report that individual level factor analysis can be explained in terms of Giant Three (or Big Five) descriptions of personality, whereas aggregate level factor analysis can be explained in terms of Gray's physiological based model. Although alternative interpretations exist, aggregate level factor analysis may correctly identify the basis of an individual's personality as a result of better reliability of measures due to aggregation. We discuss the implications of this form of analysis in terms of construct validity, personality theory, and its applicability in general. Copyright (C) 2003 John Wiley Sons, Ltd.

Drinking refusal self-efficacy questionnaire-revised (DRSEQ-R): a new factor structure with confirmatory factor analysis

The drinking refusal self-efficacy questionnaire (DRSEQ: Young, R.M., Oei, T.P.S., 1996. Drinking expectancy profile: test manual. Behaviour Research and Therapy Centre, University of Queensland, Australia Young, R.M., Oei, T.P.S., Crook, G.M., 1991. Development of a drinking refusal self-efficacy questionnaire. J. Psychopathol. Behav. Assess., 13, 1-15) assesses a person's belief in their ability to resist alcohol. The DRSEQ is a sound psychometric instrument based on exploratory factor analyses, but has not been subjected to confirmatory factor analysis. In total 2773 participants were used to confirm the factor structure of the DRSEQ. Initial analyses revealed that the original structure was not confirmed in the current study. Subsequent analyses resulted in a revised factor structure (DRSEQ-R) being confirmed in community, student and clinical samples. The DRSEQ-R was also found to have good construct and concurrent validity. The factor structure of the DRSEQ-R is more stable than the original structure of the DRSEQ and the revised scale has considerable potential in future alcohol-related research. (c) 2004 Elsevier Ireland Ltd. All rights reserved.

Confirmatory factor analysis and psychometric properties of the Anxiety Sensitivity Index-Revised in clinical and normative populations

The present investigation aimed to critically examine the factor structure and psychometric properties of the Anxiety Sensitivity Index - Revised (ASI-R). Confirmatory factor analysis using a clinical sample of adults (N = 248) revealed that the ASI-R could be improved substantially through the removal of 15 problematic items in order to account for the most robust dimensions of anxiety sensitivity. This modified scale was renamed the 21-item Anxiety Sensitivity Index (21-item ASI) and reanalyzed with a large sample of normative adults (N = 435), revealing configural and metric invariance across groups. Further comparisons with other alternative models, using multi-sample analysis, indicated the 21-item ASI to be the best fitting model for both groups. There was also evidence of internal consistency, test-retest reliability, and construct validity for both samples suggesting that the 21-item ASI is a useful assessment device for investigating the construct of anxiety sensitivity in both clinical and normative populations.

Factor analysis of the self-report Dysexecutive (DEX-S) Questionnaire

The self-rating Dysexecutive Questionnaire (DEX-S) is a recently developed standardized self-report measure of behavioral difficulties associated with executive functioning such as impulsivity, inhibition, control, monitoring, and planning. Few studies have examined its construct validity, particularly for its potential wider use across a variety of clinical and nonclinical populations. This study examines the factor structure of the DEX-S questionnaire using a sample of nonclinical (N = 293) and clinical (N = 49) participants. A series of factor analyses were evaluated to determine the best factor solution for this scale. This was found to be a 4-factor solution with factors best described as inhibition, intention, social regulation, and abstract problem solving. The first 2 factors replicate factors from the 5-factor solutions found in previous studies that examined specific subpopulations. Although further research is needed to evaluate the factor structure within a range of subpopulations, this study supports the view that the DEX has the factor structure sufficient for its use in a wider context than only with neurological or head-injured patients. Overall, a 4-factor solution is recommended as the most stable and parsimonious solution in the wider context.

Subtypes of Illicit Drug Users: A Latent Class Analysis of Data From an Australian Twin Sample

This article applies methods of latent class analysis (LCA) to data on lifetime illicit drug use in order to determine whether qualitatively distinct classes of illicit drug users can be identified. Self-report data on lifetime illicit drug use (cannabis, stimulants, hallucinogens, sedatives, inhalants, cocaine, opioids and solvents) collected from a sample of 6265 Australian twins (average age 30 years) were analyzed using LCA. Rates of childhood sexual and physical abuse, lifetime alcohol and tobacco dependence, symptoms of illicit drug abuse/dependence and psychiatric comorbidity were compared across classes using multinomial logistic regression. LCA identified a 5-class model: Class 1 (68.5%) had low risks of the use of all drugs except cannabis; Class 2 (17.8%) had moderate risks of the use of all drugs; Class 3 (6.6%) had high rates of cocaine, other stimulant and hallucinogen use but lower risks for the use of sedatives or opioids. Conversely, Class 4 (3.0%) had relatively low risks of cocaine, other stimulant or hallucinogen use but high rates of sedative and opioid use. Finally, Class 5 (4.2%) had uniformly high probabilities for the use of all drugs. Rates of psychiatric comorbidity were highest in the polydrug class although the sedative/opioid class had elevated rates of depression/suicidal behaviors and exposure to childhood abuse. Aggregation of population-level data may obscure important subgroup differences in patterns of illicit drug use and psychiatric comorbidity. Further exploration of a 'self-medicating' subgroup is needed.

Use of latent profile analysis to identify eating disorder phenotypes in an adult Australian twin cohort

Context: The relationships among the different eating disorders that exist in the community are poorly understood, especially for residual disorders in which bingeing or purging occurs in the absence of other behaviors. Objective: To examine a community sample for the number of mutually exclusive weight and eating profiles. Design: Data regarding lifetime eating disorder symptoms and weight range were submitted to a latent profile analysis. Profiles were compared regarding personality, current eating and weight, retrospectively reported life events, and lifetime depressive psychopathology. Setting: Longitudinal study among female twins from the Australian Twin Registry in whom eating was assessed by a telephone interview. Participants: A community sample of 1002 twins (individuals) who had participated in earlier waves of data collection. Main Outcome Measures: Number and clinical character of latent profiles. Results: The best fit was a 5-profile solution with women who were (1) of normal weight with few lifetime eating disorders (4.3%), (2) overweight (10.6% had a lifetime eating disorder), (3) underweight and generally had no eating disorders except for 5.3% who had restricting anorexia nervosa, (4) of low to normal weight (89.0% had a lifetime eating disorder), and (5) obese (37.0% had a lifetime eating disorder). Each profile contained more than 1 type of lifetime eating disorder except for the third profile. Women in the first and third profiles had the best functioning, with women in the fourth and fifth profiles having similarly poorer functioning. The women in the fourth group had a symptom profile distinctive from the other 4 groups in terms of severity; they were also more likely to have had lifetime major depression and suicidality. Conclusion: Lifetime weight ranges and the severity of eating disorder symptoms affected clustering more than the type of eating disorder symptom.

Evaluating the properties of a stage-specific self-efficacy scale for physical activity using classical test theory, confirmatory factor analysis and item response modeling

The purpose of this paper was to evaluate the psychometric properties of a stage-specific selfefficacy scale for physical activity with classical test theory (CTT), confirmatory factor analysis (CFA) and item response modeling (IRM). Women who enrolled in the Women On The Move study completed a 20-item stage-specific self-efficacy scale developed for this study [n = 226, 51.1% African-American and 48.9% Hispanic women, mean age = 49.2 (67.0) years, mean body mass index = 29.7 (66.4)]. Three analyses were conducted: (i) a CTT item analysis, (ii) a CFA to validate the factor structure and (iii) an IRM analysis. The CTT item analysis and the CFA results showed that the scale had high internal consistency (ranging from 0.76 to 0.93) and a strong factor structure. Results also showed that the scale could be improved by modifying or eliminating some of the existing items without significantly altering the content of the scale. The IRM results also showed that the scale had few items that targeted high self-efficacy and the stage-specific assumption underlying the scale was rejected. In addition, the IRM analyses found that the five-point response format functioned more like a four-point response format. Overall, employing multiple methods to assess the psychometric properties of the stage-specific self-efficacy scale demonstrated the complimentary nature of these methods and it highlighted the strengths and weaknesses of this scale.

Latent class and genetic analysis does not support migraine with aura and migraine without aura as separate entities

Latent class and genetic analyses were used to identify subgroups of migraine sufferers in a community sample of 6,265 Australian twins (55% female) aged 25-36 who had completed an interview based on International Headache Society UHS) criteria. Consistent with prevalence rates from other population-based studies, 703 (20%) female and 250 (9%) male twins satisfied the IHS criteria for migraine without aura (MO), and of these, 432 (13%) female and 166 (6%) male twins satisfied the criteria for migraine with aura (MA) as indicated by visual symptoms. Latent class analysis (LCA) of IHS symptoms identified three major symptomatic classes, representing 1) a mild form of recurrent nonmigrainous headache, 2) a moderately severe form of migraine, typically without visual aura symptoms (although 40% of individuals in this class were positive for aura), and 3) a severe form of migraine typically with visual aura symptoms (although 24% of individuals were negative for aura). Using the LCA classification, many more individuals were considered affected to some degree than when using IHS criteria (35% vs. 13%). Furthermore, genetic model fitting indicated a greater genetic contribution to migraine using the LCA classification (heritability, h(2) =0.40; 95% CI, 0.29-0.46) compared with the IHS classification (h(2)=0.36; 95% CI, 0.22-0.42). Exploratory latent class modeling, fitting up to 10 classes, did not identify classes corresponding to either the IHS MO or MA classification. Our data indicate the existence of a continuum of severity, with MA more severe but not etiologically distinct from MO. In searching for predisposing genes, we should therefore expect to find some genes that may underlie all major recurrent headache subtypes, with modifying genetic or environmental factors that may lead to differential expression of the liability for migraine. (C) 2004 Wiley-Liss, Inc.

Psychometric properties of the drinking expectancy questionnaire: A review of the factor structure and a proposed new scoring method

Objective: Expectancies about the outcomes of alcohol consumption are widely accepted as important determinants of drinking. This construct is increasingly recognized as a significant element of psychological interventions for alcohol-related problems. Much effort has been invested in producing reliable and valid instruments to measure this construct for research and clinical purposes, but very few have had their factor structure subjected to adequate validation. Among them, the Drinking Expectancies Questionnaire (DEQ) was developed to address some theoretical and design issues with earlier expectancy scales. Exploratory factor analyses, in addition to validity and reliability analyses, were performed when the original questionnaire was developed. The object of this study was to undertake a confirmatory analysis of the factor structure of the DEQ. Method: Confirmatory factor analysis through LISREL 8 was performed using a randomly split sample of 679 drinkers. Results: Results suggested that a new 5-factor model, which differs slightly from the original 6-factor version, was a more robust measure of expectancies. A new method of scoring the DEQ consistent with this factor structure is presented. Conclusions: The present study shows more robust psychometric properties of the DEQ using the new factor structure.

The gambling related cognitions scale (GRCS): development, confirmatory factor validation and psychometric properties

Aims The aims of this study are to develop and validate a measure to screen for a range of gambling-related cognitions (GRC) in gamblers. Design and participants A total of 968 volunteers were recruited from a community-based population. They were divided randomly into two groups. Principal axis factoring with varimax rotation was performed on group one and confirmatory factor analysis (CFA) was used on group two to confirm the best-fitted solution. Measurements The Gambling Related Cognition Scale (GRCS) was developed for this study and the South Oaks Gambling Screen (SOGS), the Motivation Towards Gambling Scale (MTGS) and the Depression Anxiety Stress Scale (DASS-2 1) were used for validation. Findings Exploratory factor analysis performed using half the sample indicated five factors, which included interpretative control/bias (GRCS-IB), illusion of control (GRCS-IC), predictive control (GRCS-PC), gambling-related expectancies (GRCS-GE) and a perceived inability to stop gambling (GRCS-IS). These accounted for 70% of the total variance. Using the other half of the sample, CFA confirmed that the five-factor solution fitted the data most effectively. Cronbach's alpha coefficients for the factors ranged from 0.77 to 0.91, and 0.93 for the overall scale. Conclusions This paper demonstrated that the 23-item GRCS has good psychometric properties and thus is a useful instrument for identifying GRC among non-clinical gamblers. It provides the first step towards devising/adapting similar tools for problem gamblers as well as developing more specialized instruments to assess particular domains of GRC.