Evaluating Misclassification Probability Using Empirical Risk

* The work is supported by RFBR, grant 04-01-00858-a

Static and Dynamic Integrated Expert Systems: State of the Art, Problems and Trends

Systemized analysis of trends towards integration and hybridization in contemporary expert systems is conducted, and a particular class of applied expert systems, integrated expert systems, is considered. For this purpose, terminology, classification, and models, proposed by the author, are employed. As examples of integrated expert systems, Russian systems designed in this field and available to the majority of specialists are analyzed.

The effect of timeliness and credit ratings on the information content of earnings announcements

Abstract: This paper investigates the impact of timeliness and credit ratings on the information content of the earnings announcements of Greek listed firms from 2001 to 2008. Using the classical event study methodology and regression analysis, we find that firms tend to release good news on time and are inclined to delay the release of bad news. We also provide evidence that the level of corporate risk differentiates the information content of earnings according to the credit rating category. Specifically, firms displaying high creditworthiness enjoy positive excess returns on earnings announcement dates. In contrast, firms with low creditworthiness undergo significant share price erosions on earnings announcement days. We also observe a substitution effect between timeliness and credit ratings in relation to the information content of earnings announcements. Specifically, we find that as the credit category of earnings-announcing firms improves, the informational role of timeliness is mitigated.

Software and Services − Research and Education in Bulgaria

Доклад по покана, поместен в сборника на Националната конференция "Образованието в информационното общество", Пловдив, май, 2010 г.

Fraction Space Revisited

Бойко Бл. Банчев - Знае се, че рационалните числа образуват интересни и богати на изчислителни възможности структури като редици на Фарей (Феъри) и безкрайни дървета. Малко внимание се обръща на по-общо, систематично излагане на основните свойства на дробите като множество. Понятия биват въвеждани без обосноваване, някои доказателства са ненужно изкуствени, а почти винаги и едните, и другите като че биват отнесени към една или друга особена структура, вместо към множеството на дробите изобщо. Изненадващо е, че някои същностни твърдения изглежда дори не са формулирани в литературата по теория на числата. Тази статия има за цел да подобри състоянието на нещата в това отношение, като предлага общо, подходящо подредено изложение на понятия и свързани с тях твърдения. Като допълнение са представени бележки върху пораждането на множеството от всички дроби – откритие значително по-старо, отколкото е прието да се смята.

Networks, space and organisational performance:a study of the determinants of industrial research income generation by universities

This paper examines the extent to which both network structure and spatial factors impact on the organizational performance of universities as measured by the generation of industrial research income. Drawing on data concerning the interactions of universities in the UK with large research and development (R&D)-intensive firms, the paper employs both social network analysis and regression analysis. It is found that the structural position of a university within networks with large R&D-intensive firms is significantly associated with the level of research income gained from industry. Spatial factors, on the other hand, are not found to be clearly associated with performance, suggesting that universities operate on a level playing field across regional environments once other factors are controlled for.

Exploring demographic information in social media for product recommendation

In many e-commerce Web sites, product recommendation is essential to improve user experience and boost sales. Most existing product recommender systems rely on historical transaction records or Web-site-browsing history of consumers in order to accurately predict online users’ preferences for product recommendation. As such, they are constrained by limited information available on specific e-commerce Web sites. With the prolific use of social media platforms, it now becomes possible to extract product demographics from online product reviews and social networks built from microblogs. Moreover, users’ public profiles available on social media often reveal their demographic attributes such as age, gender, and education. In this paper, we propose to leverage the demographic information of both products and users extracted from social media for product recommendation. In specific, we frame recommendation as a learning to rank problem which takes as input the features derived from both product and user demographics. An ensemble method based on the gradient-boosting regression trees is extended to make it suitable for our recommendation task. We have conducted extensive experiments to obtain both quantitative and qualitative evaluation results. Moreover, we have also conducted a user study to gauge the performance of our proposed recommender system in a real-world deployment. All the results show that our system is more effective in generating recommendation results better matching users’ preferences than the competitive baselines.

Education, earnings, and rates of return: The case of Taiwan

Educational planners and economists have long recognized the importance of education as a form of productive investment in both advanced countries and developing countries. In the case of Taiwan, along with impressive economic growth, there was an even faster rate of growth of the government's investment in education. This leads some to question whether education has any role in economic development. ^ The purpose of this study is to provide a broad overview of the role of education, in terms of private rates of return to education, in Taiwan. In the process, a variety of hypotheses about human capital theory are examined and an empirical study of Taiwan's earnings functions are tested to show that education can be an important instrument to increase private rate of return, even under conditions of scarcity of natural and physical resources. Data was collected using the Manpower Survey and Manpower Utilization Survey, conducted by the government. Research questions were analyzed using descriptive statistics, frequencies, and regression analysis. ^ Results indicated that the Manpower Development Plans have been the decisive influence in allowing Taiwan to develop its human resources and achieve success in meeting the needs of Taiwan's economy. The structure of age-earnings profiles showed a strong relationship between earnings and education, and the profiles that successively shift upward are associated with higher levels of education. In the cross-sectional results of the rate of return in 1997, each additional year of schooling leads to a 6.2% increase in income. As to the private rates of return to different levels of education, the results found that the private rates of return are 2.88%, 4.85% and 10.05% for primary, secondary and higher education respectively. In an intertemporal comparison for 1980, 1985, 1990 and 1997, the results showed no significant trend except the private rates of return for primary education have been falling from 3.9% to 2.88%. ^ On the basis of this study, for individual student or family in Taiwan, there is likely to be a strong demand for education, particularly at the higher level. Therefore, a well-developed higher educational level becomes essential and the content of curriculum in higher education becomes another crucial question facing planners in Taiwan if they are going to use education as a means to foster economic development. ^

The commercial bark harvest of the African cherry (Prunus africana) on Mount Oku, Cameroon: Effects on traditional uses and population dynamics

Bark extracts of the African cherry (Prunus africana) are used to treat benign prostatic hyperplasia. This study examined the effects of commercial bark harvest on population dynamics in the Kilum-Ijim Forest Preserve on Mount Oku, Cameroon and on traditional uses. P. africana is valued for its timber and as fuel although its greatest value is as a traditional medicine for human and animal ailments. Harvest has depleted the resource and has eroded traditional forest protection practices. I constructed matrix models to examine the effects of bark harvest on population structure and on population dynamics in harvested and unharvested populations. Harvesting simulations examined the effect on the population growth rate (λ) with differing levels of mortality of harvest-sized and large trees and differing harvest frequencies. Size class frequencies for the entire forest decreased in a reverse j-shaped curve, indicating adequate recruitment in the absence of harvest. Individual plots showed differences from the overall forest data, suggesting effects of natural and man-made perturbations, particularly due to bark harvest. One plot (harvested in the 1980s) showed a temporal difference in λ and fluctuated around one, due to alternating high and low fruiting years; other unharvested plots showed smaller temporal differences. Harvested plots (harvested illegally in 1997) had values of λ less than one and showed small temporal differences. The control plot also showed λ less than one, due to poor recruitment in the closed canopy forest. The value of λ for the combined data was 0.9931 suggesting a slightly declining population. The elasticity matrix for the combined data indicated the population growth rate was most sensitive to the survival of the large reproductive trees (42.5% of the elasticity). In perturbation analyses, reducing the survival of the large trees caused the largest reductions in λ. Simulations involving harvesting frequency indicated λ returns to pre-harvest conditions if trees are re-harvested after 10–15 years, but only if the large trees are left unharvested. Management scenarios suggest harvest can be sustainable if seedlings and small saplings are planted in the forest and actively managed, although large-scale plantations may be the only feasible option to meet market demand. ^

Principal component analysis and assessment of language network activation patterns in pediatric epilepsy

This dissertation establishes a novel data-driven method to identify language network activation patterns in pediatric epilepsy through the use of the Principal Component Analysis (PCA) on functional magnetic resonance imaging (fMRI). A total of 122 subjects’ data sets from five different hospitals were included in the study through a web-based repository site designed here at FIU. Research was conducted to evaluate different classification and clustering techniques in identifying hidden activation patterns and their associations with meaningful clinical variables. The results were assessed through agreement analysis with the conventional methods of lateralization index (LI) and visual rating. What is unique in this approach is the new mechanism designed for projecting language network patterns in the PCA-based decisional space. Synthetic activation maps were randomly generated from real data sets to uniquely establish nonlinear decision functions (NDF) which are then used to classify any new fMRI activation map into typical or atypical. The best nonlinear classifier was obtained on a 4D space with a complexity (nonlinearity) degree of 7. Based on the significant association of language dominance and intensities with the top eigenvectors of the PCA decisional space, a new algorithm was deployed to delineate primary cluster members without intensity normalization. In this case, three distinct activations patterns (groups) were identified (averaged kappa with rating 0.65, with LI 0.76) and were characterized by the regions of: (1) the left inferior frontal Gyrus (IFG) and left superior temporal gyrus (STG), considered typical for the language task; (2) the IFG, left mesial frontal lobe, right cerebellum regions, representing a variant left dominant pattern by higher activation; and (3) the right homologues of the first pattern in Broca's and Wernicke's language areas. Interestingly, group 2 was found to reflect a different language compensation mechanism than reorganization. Its high intensity activation suggests a possible remote effect on the right hemisphere focus on traditionally left-lateralized functions. In retrospect, this data-driven method provides new insights into mechanisms for brain compensation/reorganization and neural plasticity in pediatric epilepsy.

Secondary stakeholders as agents of influence: Three essays on political risk, reputation, and multinational performance

Organizational researchers have recently taken an interest in the ways in which social movements, non-governmental organizations (NGOs), and other secondary stakeholders attempt to influence corporate behavior. Scholars, however, have yet to carefully probe the link between secondary stakeholder legal action and target firm stock market performance. This is puzzling given the sharp rise in NGO-initiated civil lawsuits against corporations in recent years for alleged overseas human rights abuses and environmental misconduct. Furthermore, few studies have considered how such lawsuits impact a target firm’s intangible assets, namely its image and reputation. Structured in the form of three essays, this dissertation examined the antecedents and consequences of secondary stakeholder legal activism in both conceptual and empirical settings. ^ Essay One argued that conventional approaches to understanding political risk fail to account for the reputational risks to multinational enterprises (MNEs) posed by transnational networks of human rights NGOs employing litigation-based strategies. It offered a new framework for understanding this emerging challenge to multinational corporate activity. Essay Two empirically tested the relationship between the filing of human rights-related civil lawsuits and corporate stock market performance using an event study methodology and regression analysis. The statistical analysis performed showed that target firms experience a significant decline in share price upon filing and that both industry and nature of the lawsuit are significantly and negatively related to shareholder wealth. Essay Three drew upon social movement and social identity theories to develop and test a set of hypotheses on how secondary stakeholder groups select their targets for human rights-related civil lawsuits. The results of a logistic regression model offered support for the proposition that MNE targets are chosen based on both interest and identity factors. The results of these essays suggest that legal action initiated by secondary stakeholder groups is a new and salient threat to multinational business and that firms doing business in countries with weak political institutions should factor this into corporate planning and take steps to mitigate their exposure to such risks.^

Effect of delay of first interview and repetition of interview on accuracy and confidence of recall of a flashbulb -type memory

Hearing of the news of the death of Diana, Princess of Wales, in a traffic accident, is taken as an analogue for being a percipient but uninvolved witness to a crime, or a witness to another person's sudden confession to some illegal act. This event (known in the literature as a “reception event”) has previously been hypothesized to cause one to form a special type of memory commonly known as a “flashbulb memory” (FB) (Brown and Kulik, 1977). FB's are hypothesized to be especially resilient against forgetting, highly detailed including peripheral details, clear, and inspiring great confidence in the individual for their accuracy. FB's are dependent for their formation upon surprise, emotional valence, and impact, or consequentiality to the witness of the initiating event. FB's are thought to be enhanced by frequent rehearsal. FB's are very important in the context of criminal investigation and litigation in that investigators and jurors usually place great store in witnesses, regardless of their actual accuracy, who claim to have a clear and complete recollection of an event, and who express this confidently. Therefore, the lives, or at least the freedom, of criminal defendants, and the fortunes of civil litigants hang on the testimony of witnesses professing to have FB's. ^ In this study, which includes a large and diverse sample (N = 305), participants were surveyed within 2–4 days after hearing of the fatal accident, and again at intervals of 2 and 4 weeks, 6, 12, and 18 months. Contrary to the FB hypothesis, I found that participants' FB's degraded over time beginning at least as early as two weeks post event. At about 12 months the memory trace stabilized, resisting further degradation. Repeated interviewing did not have any negative affect upon accuracy, contrary to concerns in the literature. Analysis by correlation and regression indicated no effect or predictive power for participant age, emotionality, confidence, or student status, as related to accuracy of recall; nor was participant confidence in accuracy predicted by emotional impact as hypothesized. Results also indicate that, contrary to the notions of investigators and jurors, witnesses become more inaccurate over time regardless of their confidence in their memories, even for highly emotional events. ^

Participation in a positive youth development program and multi -problem alternative high school students' transformative goals: An investigation of the theoretical and methodological utility and validity of the Transformative Goal Attainment (TGAS)

There is a growing body of literature that provides evidence for the efficacy of positive youth development programs in general and preliminary empirical support for the efficacy of the Changing Lives Program (CLP) in particular. This dissertation sought to extend previous efforts to develop and preliminarily examine the Transformative Goal Attainment Scale (TGAS) as a measure of participant empowerment in the promotion of positive development. Consistent with recent advances in the use of qualitative research methods, this dissertation sought to further investigate the utility of Relational Data Analysis (RDA) for providing categorizations of qualitative open-ended response data. In particular, a qualitative index of Transformative Goals, TG, was developed to complement the previously developed quantitative index of Transformative Goal Attainment (TGA), and RDA procedures for calculating reliability and content validity were refined. Second, as a Stage I pilot/feasibility study this study preliminarily examined the potentially mediating role of empowerment, as indexed by the TGAS, in the promotion of positive development. ^ Fifty-seven participants took part in this study, forty CLP intervention participants and seventeen control condition participants. All 57 participants were administered the study's measures just prior to and just following the fall 2003 semester. This study thus used a short-term longitudinal quasi-experimental research design with a comparison control group. ^ RDA procedures were refined and applied to the categorization of open-ended response data regarding participants' transformative goals (TG) and future possible selves (PSQ-QE). These analyses revealed relatively strong, indirect evidence for the construct validity of the categories as well as their theoretically meaningful structural organization, thereby providing sufficient support for the utility of RDA procedures in the categorization of qualitative open-ended response data. ^ In addition, transformative goals (TG) and future possible selves (PSQ-QE), and the quantitative index of perceived goal attainment (TGA) were evaluated as potential mediators of positive development by testing their relationships to other indices of positive intervention outcome within a four-step method involving both analysis of variance (ANOVA and RMANOVAs) and regression analysis. Though more limited in scope than the efforts at the development and refinement of the measures of these mediators, the results were also promising. ^

Effects of computer self efficacy and positive mood in business -to -business service encounters

Attempts to improve the level of customer service delivered have resulted in an increased use of technology in the customer service environment. Customer-contact employees are expected to use computers to help them in providing better service encounters for customers. This research study done in a business-to-business environment explored the effects of customer-contact employees' computer self efficacy and positive mood on in-role customer service, extra-role customer service and organization citizenship. It also examined the relationship of customer service to customer satisfaction and customer delight. ^ Research questions were analyzed using descriptive statistics, frequency distributions, correlation analysis, and regression analysis. Results indicated that computer self efficacy had a greater impact on extra-role customer service than it did on in-role customer service. Positive mood had a positive moderating influence on extra-role customer service but not on in-role customer service. ^ There was a significant relationship between in-role customer service and customer satisfaction but not between extra-role customer service and customer satisfaction. There was no significant relationship between in-role customer service and customer delight nor between extra-role customer service and customer delight. There was a statistically greater positive relationship between joy experienced by clients and customer delight than between pleasant surprise and customer delight. ^ This study demonstrated the importance of facilitating customer-contact employee positive mood on the job in order to improve the level of extra-role customer service delivered. It also showed that increasing the level of customer service does not necessarily lead to higher levels of customer satisfaction. ^

Marl Prairie/Slough Gradients; Patterns and Trends in Shark Slough and Adjacent Marl Prairies (CERP monitoring activity 3.1.3.5), First Annual Report (2005)

The work on CERP monitoring item 3.1.3.5 (Marl prairie/slough gradients) is being conducted by Florida International University (Dr Michael Ross, Project Leader), with Everglades National Park (Dr. Craig Smith) providing administrative support and technical consultation. As of January 2006 the funds transferred by ACOE to ENP, and subsequently to FIU, have been entirely expended or encumbered in salaries or wages. The project work for 2005 started rather late in the fiscal year, but ultimately accomplished the Year 1 goals of securing a permit to conduct the research in Everglades National Park, finalizing a detailed scope of work, and sampling marsh sites which are most easily accessed during the wet season. 46 plots were sampled in detail, and a preliminary vegetation classification distinguished three groups among these sites (Sawgrass marsh, sawgrass and other, and slough) which may be arranged roughly along a hydrologic gradient from least to most persistently inundated . We also made coarser observations of vegetation type at 5-m intervals along 2 transects totaling ~ 5 km. When these data were compared with similar observations made in 1998-99, it appeared that vegetation in the western portion of Northeast Shark Slough (immediately east of the L-67 extension) had shifted toward a more hydric type during the last 6 years, while vegetation further east was unchanged in this respect. Because this classification and trend analysis is based on a small fraction of the data set that will be available after the first cycle of sampling (3 years from now), the results should not be interpreted too expansively. However, they do demonstrate the potential for gaining a more comprehensive view of marsh vegetation structure and dynamics in the Everglades, and will provide a sound basis for adaptive management.