<p>    </p> <p>Mutual learning algorithm is a machine learning algorithm where multiple machine learning algorithms learns from different sources and then share their knowledge among themselves so that all the agents can improve their classification and prediction accuracies simultaneously. Mutual learning algorithm can be an efficient mechanism for improving the machine learning and neural network efficiency in a multi-agent system. Usually, in knowledge distillation algorithms, a big network plays the role of a static teacher and passes the data to smaller networks, known as student networks, to improve the efficiency of the latter. In this thesis, it is showed that two small networks can dynamically and interchangeably play the changing roles of teacher and student to share their knowledge and hence, the efficiency of both the networks improve simultaneously. This type of dynamic learning mechanism can be very useful in mobile environment where there is resource constraint for training with big dataset. Data exchange in multi agent, teacher-student network system can lead to efficient learning.  </p>

No description available.

The inability to differentiate the English vowels /o/ and / O/ has become a stigmatized marker of a lower prestige and widespread dialect of Sri Lankan English. This lower prestige (LP) dialect is often referred to with the derogative phrase “Not pot English”. This study aims to investigate the production of the vowel contrast by native Sinhala speakers of English. To this end, speech samples of three adult learners were analyzed. The findings of the study are discussed according to hypotheses of the Speech Learning Model, which suggests that the existent L1 specific phonetic categories hinder the formation of new L2 sound categories. Here, sounds that are similar, but not identical to L1 sounds are considered to be the most difficult to acquire. Also, the percentage of L1 use and the age of second language acquisition seem to have influenced the production of the vowels. Finally, in order to address this pronunciation issue, an instructional framework to teach pronunciation is proposed.

The purpose of this study was to examine the effects of constructivist instruction on the achievement, attitude towards science teaching, science process skills and retention of fourth grade preservice science teachers in Science Teaching Methods II course. Two groups (one experimental and one control) were assigned from Hacettepe University Faculty of Education Department of Science Education. Experimental group consisted of 53 preservice science teachers and the control group consisted of 50 preservice science teachers / totally 103 preservice science teachers participated in this study. Quasi experimental research design was used in this study. Constructivist instruction was used in experimental group and traditional instruction was used in control group during the teaching and learning process. This research study was conducted in fall semester of the 2007-2008 academic year and lasted 15 weeks including the final examination term. Science Process Skills Test, Attitude towards Science Teaching Scale and Achievement Test in Science Teaching Methods II course were administered to participants three times / at the beginning of the study, immediately after the implementation process and 10 weeks later. A mixed between within ANOVA with repeated measures was used as a statistical technique for analyzing quantitative data and both descriptive and content analysis was used for analyzing questionnaire, formative and summative focus group interviews. Statistical mean difference was obtained for all tests in favor of experimental group and the findings of quantitative data analysis results were supported by the qualitative data analysis results. After interpreting the results, it can be claimed that constructivist instruction is effective in preservice science education.

The role of teacher immediacy and its impact upon student learning within the traditional classroom has been established within the instructional communication discipline in the past 30 years. In recent years, with the advent of computer-mediated distance education (i.e. online courses), some researchers have attempted to apply the same theories and measures of concepts without making the distinction between actual teacher behavioral indicants of immediacy and student perceptions of immediacy, nor recognizing that there may be a different number of variables involved between a single lesson presented online and an entire course presented over a period of time. Building upon previous models, the Short-Term Motivational Model of Learning was proposed and tested, using survey results from 229 undergraduate students who completed an online lesson presented in six different formats, and who were tested for learning outcomes. In comparison to a direct measure of learning outcomes (number of correct test answers), it was found that the Perceived Cognitive Learning Scale correlated highly with the direct measure, while the Learning Loss Scale did not. Three of the three study hypotheses were supported. Hypothesis one proposed that higher student perception of immediacy would correlate with higher student state motivation and was supported. Hypothesis two proposed that higher student trait motivation would correlate with higher student state motivation and was supported. Hypothesis three proposed that higher student state motivation would correlate with higher student cognitive learning. Student cognitive learning determined through three measures: the Perceived Cognitive Learning Scale, pretest-posttest scores differences, and the Learning Loss Scale. Using the Perceived Cognitive Learning Scale, hypothesis three was supported. Using the pretest-posttest scores differences, hypothesis three was supported weakly. Using the Learning Loss Scale, hypothesis three was also supported weakly. In testing whether teacher behavioral indicants of immediacy, student perception of immediacy and student trait motivation would explain significant variance in student state motivation in a single lesson presented online, trait motivation failed to be a predictor. Subsequently, in testing whether all of these variables would explain significant variance in student cognitive learning (and using each of the three measures of student cognitive learning), trait motivation again failed to be a predictor. In general the results support the idea that perceived learning is affected by the degree of motivation as affected by immediacy. While an attempt to vary degrees of immediacy was not fully successful, results also suggest that adding audio to online lessons may not produce significant gains in learning when compared to text-only presentations.

Made available in DSpace on 2016-03-15T19:41:21Z (GMT). No. of bitstreams: 1 Carmen Silvia de Souza Nogueira.pdf: 8637337 bytes, checksum: 138deae09a8e7a0e9889882438a0f1cf (MD5) Previous issue date: 2014-02-11 / This qualitative study analyzes the modality of learning among the families who received complaints from the school. This is done by identifying the modality of learning of the parents in order to understand their subjective position as teachers in nurturing the bond established with their children in relation to education. We assume that learning does not begin in school, but in the environment in which the child is born and raised, and many of children s attitudes outside the family environment reproduce their parents learning patterns, which in turn reveal the models they had in their relationships with their own parents. To that end, the theoretical axis focuses on authors who consider learning to be a result of not only cognitive potentiality, but also of the relationships established with a significant other, in this case with the parents as teaching figures. Our conception of learning, therefore, is that it is a binding process, shaped by the family legacy that underlies the construction of subjectivity, with each member having peculiarities in their development, albeit belonging to the same biological universe. The instruments for data collection were semi-structured interviews and graphic production of Peer Teaching and Family Teaching projection drawings from Visca (2009), amplified to measure the overlap between the current family and the parental family, thereby creating a joint assessment for the household. The data revealed the similarities and differences of parents in relation to their own and their child s learning, as well as the models brought from their own family backgrounds. The customized analysis of each family, the understanding of the place given to learning in these nuclei, the parental roles and functions associated with the children s needs, as well as the projections of the previous generation on the children, all converged with the identification of a learning model. The knowledge of family models shed light on the scope available to parents as they assist the child s achievements. By identifying parents concepts and modes of learning and the way they contribute to how children meet the challenge of school, we enable a future collaborative intervention process which takes into account points of ease and difficulty to guide the child through its education. / Esta pesquisa qualitativa buscou analisar a modalidade de aprendizagem familiar de crianças com queixa escolar, identificando a modalidade de aprendizagem dos pais e filhos para conhecer o posicionamento subjetivo dos mesmos enquanto ensinantes a nortear o vínculo estabelecido com o filho em relação ao aprender. Partimos do pressuposto de que a aprendizagem não se inicia na escola, e sim no ambiente em que a criança nasce e se desenvolve, sendo que muitas de suas atitudes no ambiente exterior ao familiar reproduzem padrões de aprendizagem dos próprios pais que, por sua vez, também revelam os modelos que tiveram em suas relações com os genitores. Para tanto o eixo teórico concentrou-se nos autores que consideram a aprendizagem como fruto não somente de potencialidades cognitivas, mas também decorrente das relações que se estabelecem com um outro significativo, neste caso com os pais, enquanto figuras ensinantes. Nossa concepção de aprendizagem, portanto, é a de que a mesma é um processo vincular, é-se herdeiro de legados familiares que perpassam a construção da subjetividade, e cada um possui peculiaridades em seu desenvolvimento, embora pertencente ao mesmo universo biológico. Os instrumentos para coleta dos dados foram entrevistas semidirigidas e a produção gráfica de desenhos projetivos Par Educativo e Família Educativa de Visca (2009), ampliando, porém, sua abrangência, ao imbricar a família atual e a de origem dos pais, fechando uma avaliação conjunta para o núcleo familiar. Os resultados encontrados permitiram verificar as semelhanças e diferenças dos pais em relação à aprendizagem deles e do filho como também os modelos trazidos de suas famílias de origem. A análise singularizada de cada família, o entendimento do lugar do saber nestes núcleos, os papéis e funções parentais associados com as demandas dos filhos bem como as projeções da geração precedente sobre as crianças confluíram na identificação de um modelo de aprendizagem. O conhecimento dos modelos familiares lançou luz quanto ao alcance que os pais têm para auxiliar nas novas conquistas do filho. Ao identificarmos as concepções e modalidades de aprendizagem dos pais e como contribuem para os enfrentamentos escolares do filho possibilitamos que um processo interventivo futuro junto a eles contemple os pontos de facilidade e dificuldade para lidar com o filho em seu processo de escolarização.

This study examines the influence of formal L2 English instruction (FEI) onpronunciation accuracy of similar and new vowels in the framework of Flege’s SpeechLearning Model (Flege, 1995a, 1999, 2002), focusing on selected segmentalpronunciation features produced by native German speakers. A spectral vowelproduction experiment was carried out with 24 German students who attend sixth, ninthand twelfth grade of a south German secondary school by the time this study wasconducted. Spectral characteristics of the similar vowel /ɛ/ and the new vowel /æ/,produced by each subject of the three groups (G6, G9, and G12), were compared tovowel data of native English speakers. In order to test the influence of formalinstruction on selected phonetic segments, which are known to cause difficulties fornative German learners of English, three native English listeners judged on a three pointscale the intelligibility of the segments produced. The results of the vowel experimentshowed significant influence of FEI on pronunciation accuracy of some characteristicsof the similar vowel /ɛ/ but not of the new vowel /æ/. The listener rating experimentshowed that FEI had on average no influence on pronunciation accuracy of the three L2English learner groups.

This study examines the influence of formal L2 English instruction (FEI) on pronunciation accuracy of similar and new vowels in the framework of Flege’s Speech Learning Model (Flege, 1995a, 1999, 2002), focusing on selected segmental pronunciation features produced by native German speakers. A spectral vowel production experiment was carried out with 24 German students who attend sixth, ninth and twelfth grade of a south German secondary school by the time this study was conducted. Spectral characteristics of the similar vowel /ɛ/ and the new vowel /æ/,produced by each subject of the three groups (G6, G9, and G12), were compared to vowel data of native English speakers. In order to test the influence of formal instruction on selected phonetic segments, which are known to cause difficulties for native German learners of English, three native English listeners judged on a three pointscale the intelligibility of the segments produced. The results of the vowel experiment showed significant influence of FEI on pronunciation accuracy of some characteristics of the similar vowel /ɛ/ but not of the new vowel /æ/. The listener rating experiment showed that FEI had on average no influence on pronunciation accuracy of the three L2 English learner groups.

This study aims at investigating the suitability of the Contrastive Analysis Hypothesis and the Speech Learning Model in describing young Swedish learners’ L2 phonology difficulties. It also explores what L1 sounds they replace L2 sounds with, and whether spelling has any influence on pronunciation. 15 Swedish students aged 9-10 were interviewed and recorded reading a word list containing minimal pairs, a text passage, and free speech. The focus was on initial and final / ʃ / and / tʃ /, initial and medial / s / and / z /, and initial / ð / and / θ /. The recordings were analysed with spectrograms and compared to a native speaker. The results were that the SLM has an advantage over the CAH. The results show that / z / was replaced by / s / 100% of the time, / tʃ / was mostly replaced by / ʃ /, but also by / k /; and that / θ / and / ð / were replaced by many different sounds that were not anticipated. The results also showed that orthography affects pronunciation for / θ /, / ð / and / tʃ /, but not for / z /. Some students had nearly established new categories for some of the new sounds, but their daily encounters with English through TV, music, and school does not seem to have had a significant part in this.

This study set out to investigate the relationship between perception and production of English / tʃ, dʒ/ and / ʒ / by native speakers of Swedish learning English in secondary school. The aim was to find out if Flege’s Speech Learning Model (SLM) (1995) and Major’s Ontogeny Phylogeny Model (OPM) (2001) are suitable to describe young learners’ L2 phonological attainment. To test perception, an identification task containing Nonwords with target sounds in initial, medial, and final position was constructed and tested on 17 school students. Three speaking tasks were carried out to test the participants’ production of the phonemes in initial, medial and final position. All participants also completed a questionnaire with questions regarding exposure and motivation. Six students were chosen for deeper analysis. The results show that both the SLM and the OPM can be used to describe young learners L2 phonology acquisition. Similar sounds were found to be harder to produce and more affected by L1 transfer, which corroborates both the SLM and the OPM. Marked sounds were affected more by universals, which was suggested by the OPM. Desire to have a native accent and reported use of the L2 seems to have effect on L2 phonology acquisition, as well. However, the relationship between all these factors is complex and the results from this study are merely suggestive.