MIDAS
Not a member yet
159 research outputs found
Sort by
Exploring Bebras Tasks Content and Performance: A Multinational Study
No full textAlthough there is no universal agreement that students should learn programming, many countries have reached a consensus on the need to expose K-12 students to Computational Thinking (CT). When, what and how to teach CT in schools are open questions and we attempt to address them by examining how well students around the world solved problems in recent Bebras challenges. We collected and analyzed performance data on Bebras tasks from 115,400 students in grades 3-12 in seven countries. Our study provides further insight into a range of questions addressed in smaller-scale inquiries, in particular about the possible impact of schools systems and gender on students' success rate.
In addition to analyzing performance data of a large population, we have classified the considered tasks in terms of CT categories, which should account for the learning implications of the challenge. Algorithms and data representation dominate the challenge, accounting for 75-90% of the tasks, while other categories such as abstraction, parallelization and problem decomposition are sometimes represented by one or two questions at various age groups. This classification can be a starting point for using online Bebras tasks to support the effective learning of CT concepts in the classroom.Although there is no universal agreement that students should learn programming, many countries have reached a consensus on the need to expose K-12 students to Computational Thinking (CT). When, what and how to teach CT in schools are open questions and we attempt to address them by examining how well students around the world solved problems in recent Bebras challenges. We collected and analyzed performance data on Bebras tasks from 115,400 students in grades 3-12 in seven countries. Our study provides further insight into a range of questions addressed in smaller-scale inquiries, in particular about the possible impact of schools systems and gender on students' success rate.
In addition to analyzing performance data of a large population, we have classified the considered tasks in terms of CT categories, which should account for the learning implications of the challenge. Algorithms and data representation dominate the challenge, accounting for 75-90% of the tasks, while other categories such as abstraction, parallelization and problem decomposition are sometimes represented by one or two questions at various age groups. This classification can be a starting point for using online Bebras tasks to support the effective learning of CT concepts in the classroom
Deep Neural Networks
No full textTyrimas, kurio tikslai yra: įvertinti giliojo mokymo neuroninių tinklų duomenų analizės algoritmų tikslumą vertinant išeminius smegenų pokyčius ir jų dinamiką trimačių galvos smegenų radiologinių tyrimų vaizduose.
Palyginti neuroninių tinklų duomenų analizės algoritmų tikslumą vertinant išeminius smegenų pokyčius su gydytojų specialistų pasiekiamais tikslumo rodikliais.
Įvertinti potencialią asistuojančio algoritmo reikšmę gerinant diagnostinio proceso kokybę ir greitį.--
COD::CIF::Parser: an error-correcting CIF parser for the Perl language
No full textA syntax-correcting CIF parser, COD::CIF::Parser, is presented that can parse CIF 1.1 files and accurately report the position and the nature of the discovered syntactic problems. In addition, the parser is able to automatically fix the most common and the most obvious syntactic deficiencies of the input files. Bindings for Perl, C and Python programming environments are available. Based on COD::CIF::Parser, the cod-tools package for manipulating the CIFs in the Crystallography Open Database (COD) has been developed. The cod-tools package has been successfully used for continuous updates of the data in the automated COD data deposition pipeline, and to check the validity of COD data against the IUCr data validation guidelines. The performance, capabilities and applications of different parsers are compared.A syntax-correcting CIF parser, COD::CIF::Parser, is presented that can parse CIF 1.1 files and accurately report the position and the nature of the discovered syntactic problems. In addition, the parser is able to automatically fix the most common and the most obvious syntactic deficiencies of the input files. Bindings for Perl, C and Python programming environments are available. Based on COD::CIF::Parser, the cod-tools package for manipulating the CIFs in the Crystallography Open Database (COD) has been developed. The cod-tools package has been successfully used for continuous updates of the data in the automated COD data deposition pipeline, and to check the validity of COD data against the IUCr data validation guidelines. The performance, capabilities and applications of different parsers are compared
Development of the Deposition Software for Small Molecule Open Crystallographic Database COD
No full textŠiame dokumente aprašomas atviros mažų molekulių kristalografinės duomenų bazės COD
automatinės duomenų įkėlimo programinės įrangos, leidžiančios užtikrinti spartų duomenų bazės
augimą, poreikis, kūrimas, kūrimo metu priimti sprendimai bei spręsti uždaviniai.This document presents the description of demand, solutions and development of the deposition software for small molecule open crystallographic database COD
Registration of publications into eLABa: interviews with VU researchers 2017
No full textDaugiausia eLABa IS per 2016 m. užregistravusių savo mokslinių publikacijų autoriai, dalyvavę 2017.04-05 mėn. tyrime apie mokslinę komunikaicją ir eLABa sistemąThe data about VU researchers were collected with the purpose to examine the factors of scientific communication that recently leads to the future developments of eLABa repository
Data file for: Governance spending fixation as a tool for reform to improve institutions
No full textMeta-data and research using only basic mathematics. File also contains description of data, sources including links to internet sites or table markers in eurostat data base and research itself that is covered by paper: Governance spending fixation as a tool for reform to improve institutionsMeta-data and research using only basic mathematics. File also contains description of data, sources including links to internet sites or table markers in eurostat data base and research itself that is covered by paper: Governance spending fixation as a tool for reform to improve institution
Deep Neural Networks
No full textTyrimas, kurio tikslai: Įvertinti giliojo mokymo neuroninių tinklų duomenų analizės algoritmų tikslumą identifikuojant ir lokalizuojant skirtingas patologijas krūtinės ląstos rentgenogramose.
Įvertinti giliojo mokymo neuroninių tinklų duomenų analizės algoritmų tikslumą automatizuotai matuojant ir skirtingus kiekybinius ir kokybinius vaizdo matavimus.
Įvertinti potencialią asistuojančio algoritmo reikšmę gerinant diagnostinio proceso kokybę ir greitį.--
Factors of Income Distribution: Labour, Capital and Welfare State
No full textProjektu siekiama ištirti kaip augant ekonomikai kinta gamybos veiksnių (darbo ir kapitalo) savininkų pajamos, kaip jos pasiskirsto skirtingos demografinės ir ekonominės struktūros namų ūkiuose, kokia valstybės institucijų įtaką gyventojų pajamoms, jų diferenciacijai. Tyrimu bandoma užpildyti iki šiol esamą spragą, kuri lieka atskirai tiriant funkcinį pajamų pasiskirstymą tarp gamybos veiksnių viso šalies ūkio lygmenyje ir atskirai namų ūkių pajamas, jų dinamiką ir nelygybę mikrolygmenyje. Susiejus šais dvi tyrimo kryptis galima atskleisti ir apibūdinti ryšius tarp ekonomikos augimo ir gyventojų pajamų.The project is aimed at analysing how income of the owners of production factors (labour and capital) change whith economic growth, how it is disributed among households of different demographic and economic structure, what is the impact of state institutions on population income and its differenciation. Proposed research project covers the gap between analysis of functional distribution of income at the state level and household income at the microlevel, its dynamics and unequal distribution. By uniting those two research strategies the links between the economic growth and income distribution can be revealed and analysed
Effects of biological traits on geographic range sizes in European amphibians
No full textThe contemporary biogeography of European amphibians has been hypothesized to have arisen from range expansion and recolonization of the northern part of the continent from southern late Pleistocene refugia, such that northern species generally possess large ranges, while southerly species are mostly small-ranged. Our aim is to test the hypothesis that these patterns are likely to be underpinned by biological traits associated with dispersal ability by means of analyses of data for anurans and urodeles, the two main groups of European amphibians.The contemporary biogeography of European amphibians has been hypothesized to have arisen from range expansion and recolonization of the northern part of the continent from southern late Pleistocene refugia, such that northern species generally possess large ranges, while southerly species are mostly small-ranged. Our aim is to test the hypothesis that these patterns are likely to be underpinned by biological traits associated with dispersal ability by means of analyses of data for anurans and urodeles, the two main groups of European amphibians
School Librarians’ Attitudes towards Teaching Information Literacy
No full textThe objective of the research is to outline school librarians’ attitudes towards teaching information literacy by analysing teaching experience, a scope of courses taught, and attitudes towards motivation of teaching. Respondents of the survey were librarians from secondary schools in Lithuania and Poland. Respondents for the research were selected from top 250 best secondary schools in Lithuania and Poland accordingly. Data for this study was gathered using Computer-assisted web Interviewing (CAWI) technique.The objective of the research is to outline school librarians’ attitudes towards teaching information literacy by analysing teaching experience, a scope of courses taught, and attitudes towards motivation of teaching. Respondents of the survey were librarians from secondary schools in Lithuania and Poland. Respondents for the research were selected from top 250 best secondary schools in Lithuania and Poland accordingly. Data for this study was gathered using Computer-assisted web Interviewing (CAWI) technique