The classification learning task requires selection of a subset of features to represent patterns to be classified. This is because the performance of the classifier and the cost of classification are sensitive to the choice of the features used to construct the classifier. Exhaustive search is impractical since it searches every possible combination of features. The runtime of heuristic and random searches are better but the problem still persists when dealing with high-dimensional datasets. We investigate a heuristic, forward, wrapper-based approach, called Linear Sequential Selection, which limits the search space at each iteration of the feature selection process. We introduce randomization in the search space. The algorithm is called Randomized Linear Sequential Selection. Our experiments demonstrate that both methods are faster, find smaller subsets and can even increase the classification accuracy. We also explore the idea of ensemble learning. We have proposed two ensemble creation methods, Feature Selection Ensemble and Random Feature Ensemble. Both methods apply a feature selection algorithm to create individual classifiers of the ensemble. Our experiments have shown that both methods work well with high-dimensional data.

As a commonly used technique in data preprocessing for machine learning, feature selection identifies important features and removes irrelevant, redundant or noise features to reduce the dimensionality of feature space. It improves efficiency, accuracy and comprehensibility of the models built by learning algorithms. Feature selection techniques have been widely employed in a variety of applications, such as genomic analysis, information retrieval, and text categorization. Researchers have introduced many feature selection algorithms with different selection criteria. However, it has been discovered that no single criterion is best for all applications. We proposed a hybrid feature selection framework called based on genetic algorithms (GAs) that employs a target learning algorithm to evaluate features, a wrapper method. We call it hybrid genetic feature selection (HGFS) framework. The advantages of this approach include the ability to accommodate multiple feature selection criteria and find small subsets of features that perform well for the target algorithm. The experiments on genomic data demonstrate that ours is a robust and effective approach that can find subsets of features with higher classification accuracy and/or smaller size compared to each individual feature selection algorithm. A common characteristic of text categorization tasks is multi-label classification with a great number of features, which makes wrapper methods time-consuming and impractical. We proposed a simple filter (non-wrapper) approach called Relation Strength and Frequency Variance (RSFV) measure. The basic idea is that informative features are those that are highly correlated with the class and distribute most differently among all classes. The approach is compared with two well-known feature selection methods in the experiments on two standard text corpora. The experiments show that RSFV generate equal or better performance than the others in many cases.

<p>Candidates’ attitudes towards various personnel selection methods get attention of organizational and work psychology specialists because of various reasons. The most important reason is that individuals’ attitudes towards personnel selection methods influence their latter behavior. Although there is a substantial amount of studies carried out in different countries, there is no data from Lithuanian and Swedish samples. The aim of current study was to analyze the attitudes towards personnel selection methods among Lithuanian students, Lithuanian employees and Swedish students. The participants (197 students and 86 employees) filled in a questionnaire, which includes short descriptions of 10 personnel selection methods as well as items about fairness of these methods. According to the results of the study, work-sample tests were ranked as the fairest personnel selection method in the Lithuanian sample. The fairest personnel selection methods in Swedish sample were work-sample tests, interview, resumes, and personal references. Lithuanian students ranked the fairness of written ability test and honesty test more favorably than Swedish students, but Swedish students tended to rank as more favorable interview, resumes, personal references and personal contacts. Personal contacts and graphology were ranked the lowest on fairness dimension in Lithuanian sample, and Swedes ranked only graphology as the least fair personnel selection method. Lithuanian employees ranked personal references, personal contacts and graphology more favorably than Lithuanian students. In Lithuanian students sample, perception of personnel selection method as a scientifically proved, logic and precise or providing an opportunity to show one’s skills, had the strongest connection with favorability ranking of personnel selection method’s fairness. In the Lithuanian employees and the Swedish students samples, perception of the method as logic or providing an opportunity to show one’s skills, had the strongest link with fairness.</p>

Candidates’ attitudes towards various personnel selection methods get attention of organizational and work psychology specialists because of various reasons. The most important reason is that individuals’ attitudes towards personnel selection methods influence their latter behavior. Although there is a substantial amount of studies carried out in different countries, there is no data from Lithuanian and Swedish samples. The aim of current study was to analyze the attitudes towards personnel selection methods among Lithuanian students, Lithuanian employees and Swedish students. The participants (197 students and 86 employees) filled in a questionnaire, which includes short descriptions of 10 personnel selection methods as well as items about fairness of these methods. According to the results of the study, work-sample tests were ranked as the fairest personnel selection method in the Lithuanian sample. The fairest personnel selection methods in Swedish sample were work-sample tests, interview, resumes, and personal references. Lithuanian students ranked the fairness of written ability test and honesty test more favorably than Swedish students, but Swedish students tended to rank as more favorable interview, resumes, personal references and personal contacts. Personal contacts and graphology were ranked the lowest on fairness dimension in Lithuanian sample, and Swedes ranked only graphology as the least fair personnel selection method. Lithuanian employees ranked personal references, personal contacts and graphology more favorably than Lithuanian students. In Lithuanian students sample, perception of personnel selection method as a scientifically proved, logic and precise or providing an opportunity to show one’s skills, had the strongest connection with favorability ranking of personnel selection method’s fairness. In the Lithuanian employees and the Swedish students samples, perception of the method as logic or providing an opportunity to show one’s skills, had the strongest link with fairness.

Žmogiškųjų išteklių vadybos magistro baigiamojo darbo tema yra aktuali, nes pagrindinis dėmesys yra skiriamas švietimo įstaigoms – gimnazijai X ir vidurinei mokyklai Y. Tarp šių institucijų egzistuoja esminis skirtumas - gimnazijose mokymas yra labiau pagilintas, moksleiviams keliami aukštesni išsilavinimo reikalavimai, suteikiamos platesnės mokymosi galimybės ir perspektyvos nei vidurinėse mokyklose. Pedagogų atrankos lyginimas ir analizė dviejose skirtingose vidurinio mokslo institucijose yra aktualus būdas palyginti ir nustatyti, kaip darbuotojų atrankos procesas skiriasi šiuose dviejuose įstaigų tipuose. Ankstesni autoriai (Aleksiūnaitė, 2009; Puslienė, 2011; Popendikytė-Stonienė, 2011; Petravičienė, 2011; Krasauskaitė, 2009), rašę darbus panašiomis temomis, gilinosi į darbuotojų atrankos specifiką kitose srityse: pašte, savivaldybėse, tarptautinėse įstaigose, atranka buvo tyrinėjama ir socialiniu aspektu. Tačiau nei vienas autorius netyrė švietimo įstaigų, nebuvo analizuojamos atskiros ir konkrečios mokslo institucijos. Būtent gimnazijų ir vidurinių mokyklų pedagogų atranka, palyginimas ir analizė yra nauja ir nepaliesta sritis. Tyrimo problema yra pateikiama teiginiu – nekokybiška, nesėkminga atranka sąlygoja prastą personalą (prastus pedagogus). Tyrimo objektas yra Vilniaus gimnazijos X ir vidurinės mokyklos Y personalas. Darbo tikslas – nustatyti, ar gimnazija X taiko efektyvesnę, kokybiškesnę atranką nei vidurinė mokykla Y. Darbe formuluojami 3 uždaviniai:... [toliau žr. visą tekstą] / The Master‘s work in Human Resource Management is relevant as an undivided attention is given to educational organisations – Gymnasium X and High School Y. There is a substantial difference between these institutions – teaching is more intense, higher educational standards for students are raised, wider educational possibilities and perspectives are offered in gymnasiums compared to high schools. The comparison and analysis of teacher selection in the two different educational institutions is a relevant way to compare and determine the difference in teacher selection processes in the two types of educational institutions. Previous authors (Aleksiūnaitė, 2009; Puslienė, 2011; Popendikytė-Stonienė, 2011; Petravičienė, 2011; Krasauskaitė, 2009) who have written works in similar subjects, have put more attention to other spheres of employee selection: posts, municipalities, international organisations, the selection process was analysed using the social aspect. There were no authors who would analyse educational institutions, there were no separate and concrete educational organisations analysed. Namely, the selection, comparison and analysis of gymnasium and high school is a new and unnafected sphere. The problem of research is presented using a proposition - ineffective, unsuccessful selection process determines poor personnel (poor teachers). The object of research is the personnel of Vilnius Gymnasium X and High School Y. The objective of research is to determine if Gymnasium... [to full text]

As part of the Accelerated Shared Growth Initiative of South Africa (AsgiSA), the South African government is focusing on improving the country's export performance (Department of Trade and Industry, 2006a). One of the objectives of the Department of Trade and Industry (DTI) is to increase the level of South African exports through the development and implementation of strategies for targeted markets. The DTI commissioned a study to scientifically identify priority products and markets in order to justify its export promotion activities and to ensure that government resources are effectively allocated. Tills led to the development of the Decision Support Model (DSM) by Viviers and Pearson (2007) as an instrument to identify realistic export opportunities for South Africa. The Viviers and Pearson (2007) study adapted and refined the methodology of Cuyvers et al. (1995) to the South African circumstances and used a sequential filtering process to identify realistic product/market export combinations for South Africa. The outcome of the DSM for South Africa was 9690 SITC product/market opportunities in seventy-four countries, clustered in twelve geographic regions. One of these regions was the Middle East. The DTI indicated the need for a study on South Africa's export opportunities to the Middle East as a result of their prioritisation of regions for export promotion. This need is therefore the rationale of this study. The research objectives were to analyse the identified Middle East countries and determine the product with the most realistic export opportunities to two of the Middle East countries and lastly, to develop a market profile to assist the DTI in promoting the exports of this product to these countries. In this study, several methods of product and market selection were investigated in order to determine the most suitable method to identify the product/market opportunities from the DSM to the Middle East. The product selection method selected involved a three-phase filtering process to determine the product with the highest export potential to the Middle East. The analysis was based on a cluster-selection process. It was determined that fruit juice was the product with the highest export potential to the region. The market selection process used a composite market potential index to determine which country in the Middle East had the highest potential for exports of fruit juice from South Africa. The two countries in the Middle East with the highest potential to import fruit juice from South Africa were Oman and Qatar. A market profile for the export of fruit juice to Oman and Qatar was developed. The market profile provided an economic overview of the two markets, analysed the market potential of fruit juice and provided technical information regarding the requirements for the export of fruit juice by South Africa to Oman and Qatar. The results of this study firstly indicate to the DTI and exporters that South Africa should export fruit juice to Oman and Qatar and secondly presents a market profile with detailed information of the process to follow in exporting fruit juice to these countries. / Thesis (M.Com. (International Commerce))--North-West University, Potchefstroom Campus, 2010.

As part of the Accelerated Shared Growth Initiative of South Africa (AsgiSA), the South African government is focusing on improving the country's export performance (Department of Trade and Industry, 2006a). One of the objectives of the Department of Trade and Industry (DTI) is to increase the level of South African exports through the development and implementation of strategies for targeted markets. The DTI commissioned a study to scientifically identify priority products and markets in order to justify its export promotion activities and to ensure that government resources are effectively allocated. Tills led to the development of the Decision Support Model (DSM) by Viviers and Pearson (2007) as an instrument to identify realistic export opportunities for South Africa. The Viviers and Pearson (2007) study adapted and refined the methodology of Cuyvers et al. (1995) to the South African circumstances and used a sequential filtering process to identify realistic product/market export combinations for South Africa. The outcome of the DSM for South Africa was 9690 SITC product/market opportunities in seventy-four countries, clustered in twelve geographic regions. One of these regions was the Middle East. The DTI indicated the need for a study on South Africa's export opportunities to the Middle East as a result of their prioritisation of regions for export promotion. This need is therefore the rationale of this study. The research objectives were to analyse the identified Middle East countries and determine the product with the most realistic export opportunities to two of the Middle East countries and lastly, to develop a market profile to assist the DTI in promoting the exports of this product to these countries. In this study, several methods of product and market selection were investigated in order to determine the most suitable method to identify the product/market opportunities from the DSM to the Middle East. The product selection method selected involved a three-phase filtering process to determine the product with the highest export potential to the Middle East. The analysis was based on a cluster-selection process. It was determined that fruit juice was the product with the highest export potential to the region. The market selection process used a composite market potential index to determine which country in the Middle East had the highest potential for exports of fruit juice from South Africa. The two countries in the Middle East with the highest potential to import fruit juice from South Africa were Oman and Qatar. A market profile for the export of fruit juice to Oman and Qatar was developed. The market profile provided an economic overview of the two markets, analysed the market potential of fruit juice and provided technical information regarding the requirements for the export of fruit juice by South Africa to Oman and Qatar. The results of this study firstly indicate to the DTI and exporters that South Africa should export fruit juice to Oman and Qatar and secondly presents a market profile with detailed information of the process to follow in exporting fruit juice to these countries. / Thesis (M.Com. (International Commerce))--North-West University, Potchefstroom Campus, 2010.

D.Litt. et Phil. / Collection development as a research topic has not been given the same prominence in the context of the public library as in that of the university library. In South Africa the future of public librarianship may well be dependent on its selection policies. The selection of fiction in public libraries has been a controversial subject since the dawn of public librarianship in the late 19th century. Librarians are still being criticised, either for their elitist attitude, because of their bias towards literary works; or for their supermarket attitude, because of their preference for popular fiction. The demand versus quality debate seems to remain unresolved. Furthermore, the role of the librarian as a developer of a particular collection, has become so confused with the role of the censor in society, that the differences between the two roles warrant an exploratory study. The aim of this study is to analyse the elements and contextual factors on a macro level that may have influence on the micro decision of selecting individual fiction titles. The study focuses on three main subject areas. Firstly, literary theory as a field of scientific research underlying the practice of literary criticism is examined to obtain perspective on the phenomenon literary quality. Secondly, the development of censorship in the Western World is reviewed as a necessary background to the interpretation of the research that has been done on reader response and reading effects in a number of disciplines. Thirdly, selection theory pertaining to the public library is explored to evaluate the amount of clarification it has brought about in major problem areas. As this is mainly an exploratory study of theories, no hypotheses have been formulated to be confirmed or rejected. Rather it must be seen as qualitative research that might generate hypotheses. The scientific aim can be described as the development of a model to visually illustrate the selection decision process and the relationships of the relevant variables within its contextual framework. Literature study, description, definition, critical analysis, contextualisation, tabulation and graphical representation, constitute the methods that have been used. The role of the selector of fiction can be clearly differentiated from that of the literary critic and the censor. It has however been found that they are interdependent because of their shared interest in the text, and because their functions are theoretically based in a body of interdisciplinary knowledge. The selector of fiction can only fulfil his role effectively when he becomes conversant with this body of knowledge. It has become clear that the public library as an open system which continually interacts with its community, has to define its own mission to provide direction and motivation to all its activities, including selection. In response to its environment the objectives of the public library must be adaptable to changes. The model which has been developed, illustrates how the selection process can resolve the problem of quality versus demand, by way of the priorities given to the phases of decision-making. It has come to the fore that there is a difference between censorship and publications control. The professional selector must know that difference and be able to handle the latter as a part of the reality of the environment in which the library operates. The research has been done from the viewpoint of Reader Studies. One of the major insights gained from the study is the close relationship between this discipline, Literature Study and Ethics, as well as the other human sciences with an interest in the interaction of readers with texts. A systems approach to literature communication, that will bring about the common involvement of these disciplines, could only be beneficial to our knowledge of readership.

The main goal of this thesis is to analyze the forms of utilizing teacher selection methods used by public high schools in the Czech Republic and in the case that shortcomings are identified, propose solutions which lead to higher validity of the selection process and to a selection of an employee, who best fulfills requirements of the specific job. The issue of employee selection is firstly defined from the theoretic perspective, further off an analysis of teacher selection methods in the Czech Republic public high school system is performed. Findings from the analysis are afterwards used for a compilation of a teacher competence model. In the conclusion an evaluation and recommendations are made in order to increase the quality of the teacher selection process on the selected school type.

No description available.