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.

A Correlation Analysis on LSA and HAL Semantic Space Models

In this paper, we compare a well-known semantic spacemodel, Latent Semantic Analysis (LSA) with another model, Hyperspace Analogue to Language (HAL) which is widely used in different area, especially in automatic query refinement. We conduct this comparative analysis to prove our hypothesis that with respect to ability of extracting the lexical information from a corpus of text, LSA is quite similar to HAL. We regard HAL and LSA as black boxes. Through a Pearsonrsquos correlation analysis to the outputs of these two black boxes, we conclude that LSA highly co-relates with HAL and thus there is a justification that LSA and HAL can potentially play a similar role in the area of facilitating automatic query refinement. This paper evaluates LSA in a new application area and contributes an effective way to compare different semantic space models.

The Use of Latent Semantic Indexing to Mitigate OCR Effects of Related Document Images

Due to both the widespread and multipurpose use of document images and the current availability of a high number of document images repositories, robust information retrieval mechanisms and systems have been increasingly demanded. This paper presents an approach to support the automatic generation of relationships among document images by exploiting Latent Semantic Indexing (LSI) and Optical Character Recognition (OCR). We developed the LinkDI (Linking of Document Images) service, which extracts and indexes document images content, computes its latent semantics, and defines relationships among images as hyperlinks. LinkDI was experimented with document images repositories, and its performance was evaluated by comparing the quality of the relationships created among textual documents as well as among their respective document images. Considering those same document images, we ran further experiments in order to compare the performance of LinkDI when it exploits or not the LSI technique. Experimental results showed that LSI can mitigate the effects of usual OCR misrecognition, which reinforces the feasibility of LinkDI relating OCR output with high degradation.

Latent class analysis of manic and depressive symptoms in a population-based sample in Sao Paulo, Brazil

Background: Current diagnostic criteria cannot capture the full range of bipolar spectrum. This study aims to clarify the natural co-segregation of manic-depressive symptoms occurring in the general population. Methods: Using data from the Sao Paulo Catchment Area Study, latent class analysis (LCA) was applied to eleven manic and fourteen depressive symptoms assessed through CIDI 1.1 in 1464 subjects from a community-based study in Sao Paulo, Brazil. All manic symptoms were assessed, regardless of presence of euphoria or irritability, and demographics, services used, suicidality and CIDI/DSM-IIIR mood disorders used to external validate the classes. Results: The four obtained classes were labeled Euthymics (EU; 49.1%), Mild Affectives (MA; 31.1%), Bipolars (BIP; 10.7%), and Depressives (DEP; 9%). BIP and DEP classes represented bipolar and depressive spectra, respectively. Compared to DEP class, BIP exhibited more atypical depressive characteristics (hypersomnia and increase in appetite and/or weight gain), risk of suicide, and use of services. Depressives had rates of atypical symptoms and suicidality comparable to oligosymptomatic MA class subjects. Limitations: The use of lay interviewers and DSM-IIIR diagnostic criteria, which are more restrictive than the currently used DSM-IV TR. Conclusions: Findings of high prevalence of bipolar spectrum and of atypical symptoms and suicidality as indicators of bipolarity are of great clinical importance, due to different treatment needs, and higher severity. Lifetime sub-affective and syndromic manic symptoms are clinically significant, arguing for the need Of revising DSM bipolar spectrum categories. (C) 2009 Elsevier B.V. All rights reserved.

Zygosity Diagnosis in the Absence of Genotypic Data: An Approach Using Latent Class Analysis

For zygosity diagnosis in the absence of genotypic data, or in the recruitment phase of a twin study where only single twins from same-sex pairs are being screened, or to provide a test for sample duplication leading to the false identification of a dizygotic pair as monozygotic, the appropriate analysis of respondents' answers to questions about zygosity is critical. Using data from a young adult Australian twin cohort (N = 2094 complete pairs and 519 singleton twins from same-sex pairs with complete responses to all zygosity items), we show that application of latent class analysis (LCA), fitting a 2-class model, yields results that show good concordance with traditional methods of zygosity diagnosis, but with certain important advantages. These include the ability, in many cases, to assign zygosity with specified probability on the basis of responses of a single informant (advantageous when one zygosity type is being oversampled); and the ability to quantify the probability of misassignment of zygosity, allowing prioritization of cases for genotyping as well as identification of cases of probable laboratory error. Out of 242 twins (from 121 like-sex pairs) where genotypic data were available for zygosity confirmation, only a single case was identified of incorrect zygosity assignment by the latent class algorithm. Zygosity assignment for that single case was identified by the LCA as uncertain (probability of being a monozygotic twin only 76%), and the co-twin's responses clearly identified the pair as dizygotic (probability of being dizygotic 100%). In the absence of genotypic data, or as a safeguard against sample duplication, application of LCA for zygosity assignment or confirmation is strongly recommended.

Patterns and transitions in substance use among young Swiss men: a latent transition analysis approach

This study investigates the potential stages of drug use. Data from the longitudinal Cohort Study on Substance Use Risk Factors were used (N = 5,116). Drug use (alcohol, tobacco, and 16 illicit drugs) over the previous 12 months was assessed at two time points. Patterns and trajectories of drug use were studied using latent transition analysis (LTA). This study's substantive contributions are twofold. First, the pattern of drug use displayed the well-known sequence of drug involvement (licit drugs to cannabis to other illicit drugs), but with an added distinction between two kinds of illicit drugs ("middle-stage" drugs: uppers, hallucinogens, inhaled drugs; and "final-stage" drugs: heroin, ketamine, GHB/GBL, research chemicals, crystal meth, and spice). Second, subgroup membership was stable over time, as the most likely transition was remaining in the same latent class.

Latent Class Analysis of Gambling Activities in a Sample of Young Swiss Men : Association with Gambling Problems, Substance Use Outcomes, Personality Traits and Coping Strategies.

The study aimed to identify different patterns of gambling activities (PGAs) and to investigate how PGAs differed in gambling problems, substance use outcomes, personality traits and coping strategies. A representative sample of 4989 young Swiss males completed a questionnaire assessing seven distinct gambling activities, gambling problems, substance use outcomes, personality traits and coping strategies. PGAs were identified using latent class analysis (LCA). Differences between PGAs in gambling and substance use outcomes, personality traits and coping strategies were tested. LCA identified six different PGAs. With regard to gambling and substance use outcomes, the three most problematic PGAs were extensive gamblers, followed by private gamblers, and electronic lottery and casino gamblers, respectively. By contrast, the three least detrimental PGAs were rare or non-gamblers, lottery only gamblers and casino gamblers. With regard to personality traits, compared with rare or non-gamblers, private and casino gamblers reported higher levels of sensation seeking. Electronic lottery and casino gamblers, private gamblers and extensive gamblers had higher levels of aggression-hostility. Extensive and casino gamblers reported higher levels of sociability, whereas casino gamblers reported lower levels of anxiety-neuroticism. Extensive gamblers used more maladaptive and less adaptive coping strategies than other groups. Results suggest that gambling is not a homogeneous activity since different types of gamblers exist according to the PGA they are engaged in. Extensive gamblers, electronic and casino gamblers and private gamblers may have the most problematic PGAs. Personality traits and coping skills may predispose individuals to PGAs associated with more or less negative outcomes.

Nonsuicidal Self-Injury and Suicidal Behavior: A Latent Class Analysis among Young Adults

Although there is a general consensus among researchers that engagement in nonsuicidal self-injury (NSSI) is associated with increased risk for suicidal behavior, little attention has been given to whether suicidal risk varies among individuals engaging in NSSI. To identify individuals with a history of NSSI who are most at risk for suicidal behavior, we examined individual variability in both NSSI and suicidal behavior among a sample of young adults with a history of NSSI (N = 439, Mage = 19.1). Participants completed self-report measures assessing NSSI, suicidal behavior, and psychosocial adjustment (e.g., depressive symptoms, daily hassles). We conducted a latent class analysis using several characteristics of NSSI and suicidal behaviors as class indicators. Three subgroups of individuals were identified: 1) an infrequent NSSI/not high risk for suicidal behavior group, 2) a frequent NSSI/not high risk for suicidal behavior group, and 3) a frequent NSSI/high risk for suicidal behavior group. Follow-up analyses indicated that individuals in the ‘frequent NSSI/high risk for suicidal behavior’ group met the clinical-cut off score for high suicidal risk and reported significantly greater levels of suicidal ideation, attempts, and risk for future suicidal behavior as compared to the other two classes. Thus, this study is the first to identity variability in suicidal risk among individuals engaging in frequent and multiple methods of NSSI. Class 3 was also differentiated by higher levels of psychosocial impairment relative to the other two classes, as well as a comparison group of non-injuring young adults. Results underscore the importance of assessing individual differences in NSSI characteristics, as well as psychosocial impairment, when assessing risk for suicidal behavior.

Meeting detection in video through semantic analysis

In this paper we present a novel approach to detect people meeting. The proposed approach works by translating people behaviour from trajectory information into semantic terms. Having available a semantic model of the meeting behaviour, the event detection is performed in the semantic domain. The model is learnt employing a soft-computing clustering algorithm that combines trajectory information and motion semantic terms. A stable representation can be obtained from a series of examples. Results obtained on a series of videos with different types of meeting situations show that the proposed approach can learn a generic model that can effectively be applied on the behaviour recognition of meeting situations.

Latent residual analysis in binary regression with skewed link

Model diagnostics is an integral part of model determination and an important part of the model diagnostics is residual analysis. We adapt and implement residuals considered in the literature for the probit, logistic and skew-probit links under binary regression. New latent residuals for the skew-probit link are proposed here. We have detected the presence of outliers using the residuals proposed here for different models in a simulated dataset and a real medical dataset.

Stigma toward schizophrenia: do all psychiatrists behave the same? Latent profile analysis of a national sample of psychiatrists in Brazil

Background: An important issue concerning the worldwide fight against stigma is the evaluation of psychiatrists’ beliefs and attitudes toward schizophrenia and mental illness in general. However, there is as yet no consensus on this matter in the literature, and results vary according to the stigma dimension assessed and to the cultural background of the sample. The aim of this investigation was to search for profiles of stigmatizing beliefs related to schizophrenia in a national sample of psychiatrists in Brazil. Methods: A sample of 1414 psychiatrists were recruited from among those attending the 2009 Brazilian Congress of Psychiatry. A questionnaire was applied in face-to-face interviews. The questionnaire addressed four stigma dimensions, all in reference to individuals with schizophrenia: stereotypes, restrictions, perceived prejudice and social distance. Stigma item scores were included in latent profile analyses; the resulting profiles were entered into multinomial logistic regression models with sociodemographics, in order to identify significant correlates. Results: Three profiles were identified. The “no stigma” subjects (n = 337) characterized individuals with schizophrenia in a positive light, disagreed with restrictions, and displayed a low level of social distance. The “unobtrusive stigma” subjects (n = 471) were significantly younger and displayed the lowest level of social distance, although most of them agreed with involuntary admission and demonstrated a high level of perceived prejudice. The “great stigma” subjects (n = 606) negatively stereotyped individuals with schizophrenia, agreed with restrictions and scored the highest on the perceived prejudice and social distance dimensions. In comparison with the first two profiles, this last profile comprised a significantly larger number of individuals who were in frequent contact with a family member suffering from a psychiatric disorder, as well as comprising more individuals who had no such family member. Conclusions: Our study not only provides additional data related to an under-researched area but also reveals that psychiatrists are a heterogeneous group regarding stigma toward schizophrenia. The presence of different stigma profiles should be evaluated in further studies; this could enable anti-stigma initiatives to be specifically designed to effectively target the stigmatizing group.