Wyniki wyszukiwania - BazTech

1

Learning a s part of knowledge exchange processes in the context of online work, based on the example of generation z employees

Albrychiewicz-Słocińska Anna, Nejman Żaneta, Varga Erika

Zeszyty Naukowe. Organizacja i Zarządzanie / Politechnika Śląska

|

2024

|

z. 191

9--26

EN

Purpose: The objective of the argument in this paper is to attempt at answering the question whether learning and knowledge exchange are the key factors determining online work preferences for Generation Z employees. Design/methodology/approach: The essence of knowledge management is that all knowledge, both explicit and tacit, accumulated by an organization becomes easily accessible to each of its members. This is important for decision-making processes and allows the organization to become more agile. Knowledge management is most often associated with modern information technologies. Thanks to them, streams of various data can be processed and analyzed in many different ways. However, in the literature there is an increasingly common attitude that more attention should be paid not only to the technological but also to the human aspect of knowledge management. The processes of knowledge exchange among employees have been subject to extensive research and studies, yet the recent years have added another thread to the discussion about the matter, i.e. a significant proportion of employees switching to the online work model. Based on the findings of the studies conducted on a group of employees representing Generation Z, the Principal Component Analysis (PCA) technique was applied to organize the factors with the highest relevance for the respondents in online work. Findings: PCA demonstrated that the components recognized as most important were those relating to knowledge transfer and their impact on employee efficiency, and on the other hand employee relations as a factor that supports the learning processes. Research limitations/implications: In order to dwell upon the underlying causes of this situation, it should be recommended to proceed with further in-depth qualitative research. Practical implications: What the research communicates to the organization is that although Generation Z members are aware of the significance of the knowledge transfer and learning processes and they understand the role of peer relations in these processes, they are unable to overcome the social barriers created by the online working system due to lack of appropriate skills. Originality/value: The paper reveals new aspects that play crucial role in shaping Generation Z attitude to online work from one side. On the other hand it also helps to design synthetic tool researching this area in the future.

2

Digital tools in teaching humanities: problem statement and development potential

Bilotcerkivcka Yuliia Oleksandrivna, Borisivna Blaga Olga, Yuriivna Ilyina Yuliia, Ivanivna Ostashchuk Vera

Zeszyty Naukowe Wyższej Szkoły Technicznej w Katowicach

|

2023

|

T. 16

63--74

EN

The article provides an analysis of the study on the use of digital tools in teaching humanities in university education, presenting arguments that confirm the significance of this topic. Due to the rapid development of digital technologies and the increasing importance of digital literacy, the potential that these tools can provide for improving the learning process in humanities is being explored. The advantages and challenges associated with the implementation of digital technologies in university education for teaching humanities are revealed. Special attention is given to the analysis of the possibilities of integrating digital tools into the educational process, uncovering their features and potential for enriching the learning process and enhancing the quality of teaching humanities. Further analysis and discussion of this topic aim to contribute to the search for optimal solutions and improving the outcomes of teaching humanities in the digital era, as well as studying future prospects for digitizing humanities education. The conclusions of the article emphasize the need for a balanced approach to the use of digital technologies, pedagogical relevance of their application, focusing on preserving interaction and communication, and involving students in active learning and collaboration.

PL

W artykule przedstawiono analizę problemu wykorzystania narzędzi cyfrowych w nauczaniu nauk humanistycznych, w edukacji uniwersyteckiej oraz przedstawiono argumenty potwierdzające znaczenie tego tematu. W związku z szybkim rozwojem technologii cyfrowych i rosnącym znaczeniem kompetencji cyfrowych, badany jest potencjał, jaki te narzędzia mogą wprowadzić, dla poprawy procesu nauczania w naukach humanistycznych. Przedstawiono korzyści i wyzwania związane z wdrażaniem technologii cyfrowych w edukację uniwersytecką, podczas nauczania nauk humanistycznych. Szczególną uwagę poświęcono analizie możliwości wdrożenia narzędzi cyfrowych w proces edukacyjny, ujawniając ich cechy i potencjał wzbogacenia procesu nauki, poprawy jakości nauczania nauk humanistycznych. Kolejne analizy i badania na ten temat mają na celu wspieranie poszukiwania optymalnych rozwiązań i poprawę wyników nauczania nauk humanistycznych w epoce cyfrowej, badanie perspektyw dla cyfryzacji edukacji humanistycznej. Wnioski artykułu podkreślają konieczność zrównoważonego podejścia do wykorzystania technologii cyfrowych, pedagogicznej celowości ich zastosowania, skupienia na zachowaniu interakcji i komunikacji, oraz zaangażowania studentów w aktywne uczenie się i współpracę.

3

Artificial intelligence in higher education industry. Just a brief introduction to complexity of an issue of future challenges

Geryk Marcin

Zeszyty Naukowe. Organizacja i Zarządzanie / Politechnika Śląska

|

2023

|

z. 172

201--217

EN

Purpose: The article was written for review purposes in order to bring the definition of artificial intelligence closer and briefly present the possibilities of its use in management and economic sciences, as well as in higher education. Design/methodology/approach: In order to obtain the desired information, the author conducted a research of the scientific papers on the relationship between higher education and artificial intelligence and extracted the most important conclusions and theories. Findings: The review of the literature allowed the author to determine that there are many applications for artificial intelligence in higher education, but it should be noted that it should always be under human control and verification. Originality/value: Apart from a brief attempt at the definition of AI and its use in higher education, the author also presents a critical perspective and possible threats, as well as proposes solutions that can regulate the ways of using artificial intelligence not only in higher education, but also in other areas of industry and social life.

4

Kształcenie w zakresie podstaw statystyki na wybranych kierunkach studiów Uniwersytetu Rzeszowskiego – analiza SWOT

Podolak Agnieszka, Kasprzyk Karina, Semenowicz Kinga, Małek Małgorzata

Polish Journal for Sustainable Development

|

2023

|

T. 27, cz. 1

47--54

PL

W artykule podjęto próbę oceny jakości kształcenia przedmiotu „Podstawy statystyki” oraz wskazano propozycje działań mogących poprawić jej efektywność przy wykorzystaniu analizy SWOT. Wyniki uzyskano dzięki współpracy ze studentami studiów pierwszego stopnia kierunków Odnawialne Źródła Energii i Gospodarka Odpadami, Logistyka w Sektorze Rolno-Spożywczym, Ochrona Środowiska oraz Rolnictwo realizowanych w Kolegium Nauk Przyrodniczych Uniwersytetu Rzeszowskiego. Uzyskane odpowiedzi wskazują na wiele pozytywnych aspektów aktualnego procesu kształcenia, podkreślając jednocześnie słabe strony oraz sygnalizując zagrożenia jakie mogą mieć miejsce w przyszłości. Otrzymane wnioski mogą być pomocne w zwiększaniu efektywności nauczania również innych przedmiotów.

EN

The article makes an attempt at assessing the quality of education in the subject "Basics of statistics" and inidicates proposals for actions that may improve its effectiveness using the SWOT analysis. The results were obtained thanks to cooperation with first-cycle students of Renewable Energy Sources and Waste Management Logistics in the Agri-Food Sector, Environmental Protection and Agriculture at the College of Natural Sciences of the University of Rzeszów. The answers obtained indicate many positive aspects of the current education process, while highlighting weaknesses and signalling threats that may occur in the future. The findings received may therefore be helpful in increasing the effectiveness of teaching other subjects as well.

5

Zintegrowany System Kwalifikacji w budownictwie

Fangrat Jadwiga, Głowacz Małgorzata, Lesiuk Remigiusz

Inżynier Budownictwa

|

2023

|

nr 7-8

28--30

6

Cloud technologies in education: the bibliographic review

Yurchenko Artem, Rozumenko Anzhela, Rozumenko Anatolii, Momot Roman, Semenikhina Olena

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2023

|

T. 13, nr 4

79--84

EN

The paper considers the use of cloud technologies in education through the prism of bibliographic analysis. The article characterizes the current state of cloud technologies in education, summarizes the trends, and forecasts the directions of recent scientific research. The leading research methods were bibliographic (visual and quantitative) analysis of keyword networks and qualitative discussion. The bibliographic analysis is based on publications indexed by the scientometric database Web Of Science over the past 20 years. The sample for analysis was formed by searching for the words cloud technology, education, learning, and teaching. The results of the study showed: a significant increase in the popularity of cloud technologies in education in recent years; an increase in the number of studies related to various aspects of educational activities under the influence of Industry 4.0; a gradual increase in the number of studies on the virtualization of the educational process and the use of artificial intelligence in education; dissemination of research on the effectiveness of various types of training using cloud services and teaching methods based on artificial intelligence; the relevance of the trend of visualization of educational material and visual analysis in education. The qualitative discussion provided grounds to identify general trends regarding future research directions.: development of mass online courses and learning technologies (immersive, the use of virtual, augmented, and mixed reality, gaming learning technologies, BYOD approach); further virtualization of universities; development of inclusive education, educational analytics, and assessment (formative and adaptive computer assessment); early training of teachers to use cloud technologies and specialized services in subject learning; research related to visualization (big data, design, simulation, simulation of various processes, etc.) and the designing of relevant new academic disciplines; research of STEM and STEAM education.

PL

W artykule omówiono wykorzystanie technologii chmurowych w edukacji pod kątem analizy bibliograficznej. W artykule scharakteryzowano obecny stan wykorzystania technologii chmurowych w edukacji, podsumowano trendy i prognozy kierunków odpowiednich badań naukowych. Wiodącymi metodami badawczymi były wizualna i ilościowa analiza sieci słów kluczowych oraz dyskusja jakościowa. Analiza bibliograficzna została przeprowadzona na publikacjach indeksowanych przez scjentometryczną bazę Web Of Science przez ostatnie 20 lat. Próbkę do analizy tworzy się poprzez wyszukiwanie słów words cloud technology, education, learning, teaching. Wyniki analizy wykazały: znaczący wzrost popularności technologii chmurowych w edukacji w ostatnich latach; wzrost liczby badań związanych z różnymi aspektami działalności edukacyjnej pod wpływem Przemysłu 4.0; stopniowy wzrost liczby badań nad wirtualizacją procesu edukacyjnego i wykorzystaniem sztucznej inteligencji w edukacji; upowszechnianie badań nad efektywnością różnego rodzaju szkoleń z wykorzystaniem usług chmurowych oraz metod nauczania opartych na sztucznej inteligencji; znaczenie trendu wizualizacji materiałów edukacyjnych i analizy wizualnej w edukacji. Dyskusja jakościowa dała podstawy do przewidywania kierunków odpowiednich badań: rozwoju masowych kursów online i technologii uczenia się (immersyjne, wykorzystanie rzeczywistości wirtualnej, rozszerzonej i mieszanej, technologie uczenia się w grach, podejście BYOD); dalsza wirtualizacja uczelni; rozwój uczenia się włączającego, analityki edukacyjnej i oceny (formatywna i adaptacyjna ocena komputerowa); proaktywne szkolenie nauczycieli w zakresie korzystania z technologii chmurowych i specjalistycznych usług w zakresie uczenia się przedmiotów; badania związane z wizualizacją (big data, projektowanie, symulacja, symulacja różnych procesów itp.) oraz rozwój nowych dyscyplin akademickich do prezentacji różnych danych; badanie edukacji STEM i STEAM.

7

Studying „management” as a challenge to develop your own competences

Jarecki Wojciech

Zeszyty Naukowe. Organizacja i Zarządzanie / Politechnika Śląska

|

2022

|

z. 159

135--142

EN

Purpose: The first aim of the article was to investigate whether high school graduates who want to study management take up gainful employment before starting their studies and how much was their salary. The second goal was to determine the reasons for starting studies and choosing a field of study, and the third was to determine the degree of identification of professional interests of first-year students in the field of Management. Design/methodology/approach: For the research presented in this article, an analysis of the literature in the area of professional competences and career development was used. Research was also carried out using questionnaire forms with open questions. The research covered all first-year students of Management at the University of Szczecin. Findings: Almost all first-year management students worked before starting their studies. They started their studies guided mainly by their interests in the field of management and the desire to run their own business, but they do not have exactly what exactly they would like to do after graduation. Research limitations/implications: In the future, research will be continued on a larger research sample. Practical implications: students have poorly identified interests. Their task is to try to identify these interests, but the support of lecturers is also important. Originality/value: Reasons for starting studies and choosing the field of Management and Identification of students' interests. The research results can be used by academics and students.

8

Co wiemy, a co powinniśmy wiedzieć o wentylacji

Wargocki Paweł

Rynek Instalacyjny

|

2022

|

Nr 1-2

42--50

PL

Wentylacja jest bez wątpienia uznawana za podstawowy sposób zapewnienia dobrej jakości powietrza w pomieszczeniach. Chociaż szeroko opisano ją w literaturze naukowej, wciąż istnieje kilka nie do końca rozwiązanych kwestii dotyczących wentylacji. Należą do nich m.in.: Ile wentylacji potrzeba w danym budynku?; Jakie kryteria należy zastosować do wyznaczenia wentylacji?; Ile wynosi absolutne minimum dla strumienia wentylacyjnego w danym budynku?; Czy możemy wykorzystać dane epidemiologiczne do ustalenia wymagań dotyczących wentylacji? Czy wentylację można stosować jako miarę jakości powietrza w pomieszczeniach? W artykule zawarto krótki przegląd literatury dotyczący wentylacji w budynkach, stanowiący punkt wyjścia do dyskusji nad powyższymi kwestiami. Przedstawiono również historyczny rys rozwoju wymagań dotyczących wentylacji, jak i wymagań dotyczących utrzymania instalacji wentylacyjnych. Podano także kilka sugestii, jak określać wymagania dotyczące wentylacji.

EN

Ventilation is without any doubt recognized as an essential means of providing good indoor air quality. Although described widely in the scientific literature, there are still a few incompletely resolved questions concerning ventilation. They include, among others: How much ventilation is needed in a given building?; Which criteria should be used to determine ventilation?; What is the absolute minimum ventilation rate in a given building?; Can we use epidemiological data for setting ventilation requirements?; and Can ventilation be used as an indoor air quality metric? This short article presents short review on ventilation in buildings which create reference for the subsequent discussion of the listed questions. Historical view of the development of ventilation and ventilation requirements is presented as well, and the requirements regarding maintenance of systems delivering ventilation air. Some suggestions on how to determine ventilation requirements are provided.

9

On Minimization and Learning of Deterministic ω-Automata in the Presence of Don't Care Words

Löding Christof, Stachon Max Philip

Fundamenta Informaticae

|

2022

|

Vol. 189, nr 1.

69--90

EN

We study minimization problems for deterministic ω-automata in the presence of don't care words. We prove that the number of priorities in deterministic parity automata can be efficiently minimized under an arbitrary set of don't care words. We derive that from a more general result from which one also obtains an efficient minimization algorithm for deterministic parity automata with informative right-congruence (without don't care words). We then analyze languages of don't care words with a trivial right-congruence. For such sets of don't care words it is known that weak deterministic Büchi automata (WDBA) have a unique minimal automaton that can be efficiently computed from a given WDBA (Eisinger, Klaedtke 2006). We give a congruence-based characterization of the corresponding minimal WDBA, and show that the don't care minimization results for WDBA do not extend to deterministic ω-automata with informative right-congruence: for this class there is no unique minimal automaton for a given don't care set with trivial right congruence, and the minimization problem is NP-hard. Finally, we extend an active learning algorithm for WDBA (Maler, Pnueli 1995) to the setting with an additional set of don't care words with trivial right-congruence.

10

Selective maintenance optimization with stochastic break duration based on reinforcement learning

Liu Yilai, Qian Xinbo

Eksploatacja i Niezawodność

|

2022

|

Vol. 24, no. 4

771--784

EN

For industrial and military applications, a sequence of missions would be performed with a limited break between two adjacent missions. To improve the system reliability, selective maintenance may be performed on components during the break. Most studies on selective maintenance generally use minimal repair and replacement as maintenance actions while break duration is assumed to be deterministic. However, in practical engineering, many maintenance actions are imperfect maintenance, and the break duration is stochastic due to environmental and other factors. Therefore, a selective maintenance optimization model is proposed with imperfect maintenance for stochastic break duration. The model is aimed to maximize the reliability of system successfully completing the next mission. The reinforcement learning(RL) method is applied to optimally select maintenance actions for selected components. The proposed model and the advantages of the RL are verified by three case studies verify.

11

An adaptive PC-Kriging method for time-variant structural reliability analysis

Nan Hang, Li Hongshuang, Song Zhuocheng

Eksploatacja i Niezawodność

|

2022

|

Vol. 24, no. 3

532--543

EN

The practical application of time-variant reliability analysis is limited by its computationally expensive models which describe the structural system behavior. This paper presents a new adaptive PC-Kriging (APCK) approach to accurately and efficiently assess the time-variant reliabilities. Time interval is firstly discretized with a series of time instants and then the stochastic process is reconstructed by standard normal random variables and deterministic function of time. PC-Kriging (PCK) models are built at each time instant to predict the instantaneous responses of performance function. To improve the accuracy and efficiency, a new update strategy based on the integration of U- and H- learning functions is developed to refine the PCK models of instantaneous responses. One or two best samples are identified by the proposed learning criterion for updating the PCK models. Finally, Monte Carlo simulation (MCS) is used to estimate the time-variant reliability based on the updated PCK models. Four examples are used to validate the accuracy and efficiency of the proposed method.

12

The state as a learning organization and cybersecurity

Szreniawski Piotr

Cybersecurity and Law

|

2022

|

Vol. 8 (2)

36--43

EN

In a contemporary state, cybersecurity is becoming an increasingly important issue. In order for a state to adequately respond to changing threats, it must be a learning organization. Learning about cybersecurity should include specialist knowledge, but also knowledge about cybersecurity shaping the behaviours of the general public and policy makers. Legal regulations are an important instrument of shaping cybersecurity in a learning state.

PL

We współczesnym państwie cyberbezpieczeństwo staje się coraz ważniejszym zagadnieniem. Żeby państwo odpowiednio reagowało na zmieniające się zagrożenia, musi być organizacją uczącą się. Uczenie się cyberbezpieczeństwa powinno obejmować wiedzę specjalistyczną, a także wiedzę o cyberbezpieczeństwie kształtującą zachowania ogółu społeczeństwa i decydentów politycznych. Ważnym instrumentem kształtowania cyberbezpieczeństwa w państwie uczącym się są przepisy prawa.

13

Testowanie użyteczności oprogramowania do nauki języków klasycznych : case study aplikacji mobilnej Speak Latin : Learn Latin Language Offline

Zbądzki Jakub, Rogowska-Wandowicz Anna

Biuletyn Naukowy Wrocławskiej Wyższej Szkoły Informatyki Stosowanej. Informatyka

|

2022

|

vol. 9, nr 1

25--30

PL

Celem artykułu jest przedstawienie znaczenia, jakie mają kategoria użyteczności i jej testowanie dla oprogramowania służącego do nauki języków klasycznych, a także wskazanie możliwych defektów, które mogą się pojawiać, i ich źródeł. Jako przykład posłużyła aplikacja mobilna Speak Latin: Learn Latin Language Offline. Testy użyteczności przeprowadzono na aplikacji w wersji 1.0.18, w brytyjskiej wersji językowej, na urządzeniu Huawei Mate 10 Lite z systemem Android w wersji 8.1.0. Oprogramowanie testowano pod kątem efektywności i satysfakcji z użytkowania. Zaobserwowano pojawianie się istotnych defektów w kluczowych obszarach aplikacji. Usterki dotyczą przede wszystkim: 1. nieprawidłowej wymowy wyrazów; 2. braku deklarowanego rozróżniania zapisu i wymowy; 3. schematu odtwarzania nagrań istotnie utrudniającego naukę; 4. niepoprawnych i niejednolitych tłumaczeń lub ich braku; 5. nieczytelnych kategorii, w których występują dane słowa. Zauważono, że istotna część defektów wynika z zastosowania syntezatora i translatora mimo słabego rozwinięcia tych technologii w kontekście języków klasycznych. Aplikacja prawdopodobnie nie przeszła też testów użyteczności. Kontrola ze strony człowieka i weryfikacja jakości tego rodzaju oprogramowania okazują się niezbędne. Bez nich nie może ono spełniać wymagań przeciętnych odbiorców, które dotyczą efektywności i satysfakcji z użytkowania, nie jest też w stanie przynosić istotnych korzyści edukacyjnych.

EN

The aim of this article is to present the importance of the usability category and its testing for software designed for learning classical languages, as well as to point out possible defects that may appear and their sources. The mobile application Speak Latin: Learn Latin Language Offline was used as an example. Usability testing was carried out on the app version 1.0.18, in the British English language version, on a Huawei Mate 10 Lite device with the Android system in version 8.1.0. The software was tested in terms of effectiveness and satisfaction of use. Significant defects have been observed in key areas of the application. The defects relate primarily to: 1. incorrect pronunciation of words; 2. lack of claimed differentiation between notation and pronunciation; 3. the playback pattern of recordings significantly impeding learning; 4. incorrect and inconsistent translations or lack thereof; 5. illegible categories in which words occur. It was noted that a significant proportion of the defects are due to the use of synthesizer and translator despite the poor development of these technologies in the context of classical languages. The application probably did not pass usability testing either. Human control and quality verification of this kind of software prove essential. Without them, it cannot meet the average user's requirements regarding the effectiveness and satisfaction of use, nor is it able to deliver significant educational benefits.

14

Development and implementation of an online platform for curriculum mapping in medical education

Majerník Jaroslav, Kacmarikova Andrea, Komenda Martin, Kononowicz Andrzej A., Kocurek Anna, Stalmach-Przygoda Agata, Balcerzak Łukasz, Hege Inga, Ciureanu Adrian

Bio-Algorithms and Med-Systems

|

2022

|

Vol. 18, no. 1

1--11

EN

Objectives: Nowadays universities face ever-increasing demands on quality of education, which is crucial from perspective of future graduates. In face of the need of constant quality improvements of medical curricula, it is important to seek strategies for their efficient management. The general trend is to develop electronic support tools to streamline the curricular design, analysis and harmonization. Methods: Based on the requirements we have identified by the needs analysis among curriculum designers, teachers and managers at five universities involved in the Building Curriculum Infrastructure in Medical Education (BCIME) project, and evidence published in literature on curriculum development, we have developed methodological guidelines on curriculum innovations and a software-based tools that help manage, map and analyse curricula in the medical and healthcare study fields. Results: In this paper, we share our experiences with building and implementation of EDUportfolio, an online platform developed within our consortium and intended to facilitate harmonisation and optimisation of medical outcome-based curricula. Its functionalities and outputs were verified by pilot mapping of Anatomy curricula as taught at partner universities in five European countries. Conclusions: The visualisation and the analysis of described curriculum data using natural language processing techniques revealed both the hidden relations between curriculum building blocks and a set of overlaps and gaps in curricula. In addition, we demonstrate both the usability of the platform in the context of the involved academic environments and the capability to map and compare curricula across different institutions and different countries.

15

Creating awareness and implementing Blooms taxonomy in teaching pedagogy-with special reference to higher education

Agarwal Shivani, Solanki Vijender Kumar

Annals of Computer Science and Information Systems

|

2022

|

Vol. 34

75--79

EN

The main objective of this study is to design a questionnaire which helps to measure the constructive alignment of Blooms taxonomy as per the six level of cognition. To achieve the objective, questionnaire was designed. This questionnaire will be helpful in ascertaining the present level of respondent as per the blooms taxonomy so that corrective actions could be taken to optimize learning outcomes. In Indian higher education, especially in Tier 2 and Tier 3 colleges are more heavily biased towards lower order thinking skills. With this questionnaire, current standing of the respondent can be measured and progression to higher order thinking skills can be facilitated at the end of the trainer/educator, hence bridging the gap between academia and industry.

16

Convolutional neural networks training for autonomous robotics

Lozhkin Alexander, Maiorov Konstantin, Bozek Pavol

Management Systems in Production Engineering

|

2021

|

nr 1 (29)

75--79

EN

The article discusses methods for accelerating the operation of convolutional neural networks for autonomous robotics learning. The analysis of the theoretical possibility of modifying the neural network learning mechanism is carried out. Classic semiotic analysis and the theory of neural networks is proposed to union. An assumption is made about the possibility of using the symmetry mechanism to accelerate the training of convolutional neural networks. A multilayer neural network to represent how space is an attempt has been made. The conclusion was based on the laws on the plane obtained earlier. The derivation of formulas turned out to be impossible due to the problems of modern mathematics. A new approach is proposed, which involves combining the gradient descent algorithm and the stochastic completion of convolutional filters by the principles of symmetries. The identified algorithms allow increasing the learning rate from 5% to 15%, depending on the problem that the neural network solves.

17

Typology emerges from simplicity in representations and learning

Lambert Dakotah, Rawski Jonathan, Heinz Jeffrey

Journal of Language Modelling

|

2021

|

Vol. 9, No. 1

151--194

EN

We derive well-understood and well-studied subregular classes of formal languages purely from the computational perspective of algorithmic learning problems. We parameterise the learning problem along dimensions of representation and inference strategy. Of special interest are those classes of languages whose learning algorithms are necessarily not prohibitively expensive in space and time, since learners are often exposed to adverse conditions and sparse data. Learned natural language patterns are expected to be most like the patterns in these classes, an expectation supported by previous typological and linguistic research in phonology. A second result is that the learning algorithms presented here are completely agnostic to choice of linguistic representation. In the case of the subregular classes, the results fall out from traditional model-theoretic treatments of words and strings. The same learning algorithms, however, can be applied to model-theoretic treatments of other linguistic representations such as syntactic trees or autosegmental graphs, which opens a useful direction for future research.

18

Investigating the effects of i-complexity and e-complexity on the learnability of morphological systems

Johnson Tamar, Gao Kexin, Smith Kenny, Rabagliati Hugh, Culbertson Jennifer

Journal of Language Modelling

|

2021

|

Vol. 9, No. 1

97--150

EN

Research on cross-linguistic differences in morphological paradigms reveals a wide range of variation on many dimensions, including the number of categories expressed, the number of unique forms, and the number of inflectional classes. However, in an influential paper, Ackerman and Malouf (2013) argue that there is one dimension on which languages do not differ widely: in predictive structure. Predictive structure in a paradigm describes the extent to which forms predict each other, called i-complexity. Ackerman and Malouf (2013) show that although languages differ according to measure of surface paradigm complexity, called e-complexity, they tend to have low i-complexity. They conclude that morphological paradigms have evolved under a pressure for low i-complexity. Here, we evaluate the hypothesis that language learners are more sensitive to i-complexity than e-complexity by testing how well paradigms which differ on only these dimensions are learned. This could result in the typological findings Ackerman and Malouf (2013) report if even paradigms with very high e-complexity are relatively easy to learn, so long as they have low i-complexity. First, we summarize a recent work by Johnson et al. (2020) suggesting that both neural networks and human learners may actually be more sensitive to e-complexity than i-complexity. Then we build on this work, reporting a series of experiments which confirm that, indeed, across a range of paradigms that vary in either e- or icomplexity, neural networks (LSTMs) are sensitive to both, but show a larger effect of e-complexity (and other measures associated with size and diversity of forms). In human learners, we fail to find any effect of i-complexity on learning at all. Finally, we analyse a large number of randomly generated paradigms and show that e- and i-complexity are negatively correlated: paradigms with high e-complexity necessarily show low i-complexity. We discuss what these findings might mean for Ackerman and Malouf’s hypothesis, as well as the role of ease of learning versus generalization to novel forms in the evolution of paradigms.

19

Approaching explanatory adequacy in phonology using Minimum Description Length

Rasin Exer, Berger Iddo, Lan Nur, Shefi Itamar, Katzir Roni

Journal of Language Modelling

|

2021

|

Vol. 9, No. 1

17--66

EN

A linguistic theory reaches explanatory adequacy if it arrives at a linguistically-appropriate grammar based on the kind of input available to children. In phonology, we assume that children can succeed even when the input consists of surface evidence alone, with no corrections or explicit paradigmatic information – that is, in learning from distributional evidence. We take the grammar to include both a lexicon of underlying representations and a mapping from the lexicon to surface forms. Moreover, this mapping should be able to express optionality and opacity, among other textbook patterns. This learning challenge has not yet been addressed in the literature. We argue that the principle of Minimum Description Length (MDL) offers the right kind of guidance to the learner – favoring generalizations that are neither overly general nor overly specific – and can help the learner overcome the learning challenge. We illustrate with an implemented MDL learner that succeeds in learning various linguistically-relevant patterns from small corpora.

20

Learning novelty detection outside a class of random curves with application to COVID-19 growth

Rafajłowicz Wojciech

Journal of Artificial Intelligence and Soft Computing Research

|

2021

|

Vol. 11, No. 3

195--215

EN

Let a class of proper curves is specified by positive examples only. We aim to propose a learning novelty detection algorithm that decides whether a new curve is outside this class or not. In opposite to the majority of the literature, two sources of a curve variability are present, namely, the one inherent to curves from the proper class and observations errors’. Therefore, firstly a decision function is trained on historical data, and then, descriptors of each curve to be classified are learned from noisy observations.When the intrinsic variability is Gaussian, a decision threshold can be established from T2 Hotelling distribution and tuned to more general cases. Expansion coefficients in a selected orthogonal series are taken as descriptors and an algorithm for their learning is proposed that follows nonparametric curve fitting approaches. Its fast version is derived for descriptors that are based on the cosine series. Additionally, the asymptotic normality of learned descriptors and the bound for the probability of their large deviations are proved. The influence of this bound on the decision threshold is also discussed.The proposed approach covers curves described as functional data projected onto a finite-dimensional subspace of a Hilbert space as well a shape sensitive description of curves, known as square-root velocity (SRV). It was tested both on synthetic data and on real-life observations of the COVID-19 growth curves.