检索结果-内蒙古大学图书馆

Data Preprocessing with IRM for skill extraction using NMF

Procedia Computer Science 2022年 207卷 2283-2290页

作者： Shinichi Oeda Shunsuke Iizumi Department of Information and Computer Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan Advanced Course of Control and Information Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan

To provide examinees with appropriate question items, identifying and evaluating users' latent and acquired skills is important in e-learning systems. These latent skills are defined in terms of a Q-matrix. Constructing Q-matrices manually is relatively labor-intensive, because doing so requires expert insight as well as verbal examination to identify which skills students have mastered. Methods to extract a Q-matrix from examination results automatically using nonnegative matrix factorization (NMF) have been explored, and improved methods such as online NMF have been proposed, which extracts an immutable Q-matrix from time-series data. In this study, we propose a data preprocessing method to improve the accuracy of NMF using a co-clustering method called an infinite relational model (IRM). We also describe experiments conducted using synthetic data, in which learners' skills and the bias in the number of their responses were evaluated, and show that the results demonstrate the efficacy of the proposed approach.

关键词： NMF(Non-negative Matrix Factorization) IRM(Infinite Relational Model) Co-clustering Q-matrix Student-Modeling Skill-Modeling

来源：评论

学校读者我要写书评

暂无评论

Verification of Usefulness of Student Modeling with Real Educational Data using Convex Factorization Machines

引用

Procedia Computer Science 2021年 192卷 804-811页

作者： Shinichi Oeda Daiki Shimizu Department of Information and Computer Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan Advanced Course of Control and Information Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan

Extracting useful information generated from educational settings involves the application of data mining, machine learning, and statistics to the large amount of electronic data collected by educational systems. To generate better higher learning outcomes using an intelligent tutoring system, such as an e-learning system, it is necessary to more accurately understand the state of student knowledge. The purpose of student modeling is to estimate the students’ skills from log data, such as examination results, and to predict whether or not a student will be able to solve a problem. In this study, we propose a student performance prediction method using convex factorization machines. Factorization machines offer a combination of the advantages of support vector machines and factorization models such as matrix factorization. The results of conventional methods, which predict student performance using factorization machines, exhibited better results than they have before. However, because factorization machines are not convex optimizations, they acquire local minima, which is a disadvantage. Therefore, we used convex factorization machines in order to improve the performance of student modeling predictions.

关键词： Educational Data Mining Student Modeling Convex Factorization Machines Factorization Machines Convex Optimization Recommendation Intelligent Tutoring Systems

来源：评论

学校读者我要写书评

暂无评论

Latent State Estimation Method for Students Using Knowledge Tracing Combining with IRT

引用

Procedia Computer Science 2022年 207卷 1992-1999页

作者： Shinichi Oeda Toma Kakizaki Department of Information and Computer Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan Advanced Course of Control and Information Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan

The Intelligent Tutoring Systems (ITSs) are a system that provides an efficient learning environment by assigning questions that are suitable to the learner's skill state. To improve the performance of ITSs, conventional studies have proposed student models that can estimate skill states with high accuracy. However, these models were based on the assumption that the log data was correct. In other words, they do not take into account the possibility of students cheating or memorizing answers, assuming that the test was conducted fairly and impartially. In recent years, many examinations have been conducted online as a countermeasure against COVID-19 (coronavirus) infections, and students may be able to obtain hints for the examinations using the internet or books. In this way, it is important to know how students approach learning in order to estimate their skill state. In this study, we propose a method for estimating the latent state of learners using a model that combines Knowledge Tracing, the de-facto standard for student modeling method, and Item Response Theory.

关键词： Intelligent Tutoring System Educational Data Mining Student Modeling Knowledge Tracing Item Response Theory

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-based Source Code Classification and Visualization of Decision Rationales

引用

Procedia Computer Science 2024年 246卷 2753-2761页

作者： Shinichi Oeda Seiya Mitsumori Department of Information and Computer Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan Advanced Course of Control and Information Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan

In programming education, it is crucial to provide instruction that is tailored to students’ proficiency levels. For this purpose, an objective evaluation of each student’s coding ability is essential. Additionally, it is necessary to understand the characteristics of source code written by both beginners and advanced students. Previous research has successfully converted the structural information of source code into graphs and assessed coding skills using deep learning, achieving high accuracy. However, it remains unclear which specific structural elements significantly influence these assessments. This study addresses this gap by transforming source code into abstract syntax trees and developing a model that uses Graph Convolutional Networks to categorize code as either beginner or advanced users based on learned structural information. Furthermore, we apply Integrated Gradients to visualize the decision-making basis of our model and elucidate the structural characteristics distinguishing source code written by beginner and advanced users.

关键词： Abstract Syntax Tree Deep Learning Explainable AI Graph Convolutional Networks Integrated Gradients Source Code Classification

来源：评论

学校读者我要写书评

暂无评论

Visualization of Programming Skill Structure by Log-Data Analysis with Decision Tree

引用

Procedia Computer Science 2019年 159卷 582-589页

作者： Shinichi Oeda Mutsumi Chieda Department of Information and Computer Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan Advanced Course of Control and Information Engineering National Institute of Technology Kisarazu College 11-1 Kiyomidaihigashi 2-chome Kisarazu City Chiba Japan

To evaluate the ability of a programmer in universities or staff agencies, evaluators conduct examination in which applicants solve problems, or they refer to the applicants’ GitHub and review their code. However, the collected source code is evaluated by a human under present conditions. This leads to two problems: many source codes cannot be evaluated simultaneously, and it is difficult to maintain consistency in the evaluation criteria among evaluators. We propose methods to estimate the current skill of the programmer, which can be used by the evaluators to understand an applicant’s skill by analyzing the collected source code automatically. In particular, this study visualizes the feature of the students’ source code in term of quarters using a decision tree to estimate the programming skill.

关键词： Programming lesson Beginner Programmer Decision Tree Random Forest Latent Skill EDM(Educational Data Mining)

来源：评论

学校读者我要写书评

暂无评论

Fabrication of diamond nanopit arrays by room-temperature curing nanoimprint lithography using glass-like carbon molds

Fabrication of diamond nanopit arrays by room-temperature cu...

引用

2011 MRS Fall Meeting

作者： Kiyohara, Shuji Ito, Chigaya Ishikawa, Ippei Takikawa, Hirofumi Taguchi, Yoshio Sugiyama, Yoshinari Omata, Yukiko Kurashima, Yuichi Electric and Control System Engineering Course Faculty of Advanced Engineering Maizuru National College of Technology 234 Aza Shiroya Maizuru Kyoto 625-8511 Japan Department of Electrical and Electronic Information Engineering Toyohashi University of Technology Toyohashi Aichi 441-8580 Japan Application and Technical Section Elionix Inc. 3-7-6 Motoyokoyama Hachioji Tokyo 192-0063 Japan Department of Mechanical System Engineering University of Yamanashi 4-3-11 Takeda Kofu Yamanashi 400-8511 Japan

ISBN: (纸本)9781605113722

We have proposed the use of glass-like carbon (GC), as mold material because the 27-maximum etching selectivity of polysiloxane film against GC, which was approximately six times larger than that of polysiloxane film against chemical vapor deposited (CVD) diamond film. We have investigated the fabrication of diamond nanopit arrays by room-temperature curing nanoimprint lithography (RTC-NIL) using GC mold, as applications to the emitter and the micro-gear. The polysiloxane has in the state of sticky liquid at room-temperature and negative-exposure characteristic. Therefore, the polysiloxane was used as RTC-imprint resist material, and also used as electron beam (EB) resist (oxide mask) material in EB lithography. We have fabricated the cylindrical GC nanodot mold with 500 nm-diameter, 600 nm-height and 2 μm-pitch. We carried out RTC-NIL using GC mold under the following optimum conditions: time from spin-coating to imprint of 1 min, imprinting pressure of 0.5 MPa and imprinting time of 5 min. Then, we have processed the diamond film with an electron cyclotron resonance (ECR) oxygen ion shower. We have fabricated diamond nanopit array with 250 nm-depth and 500 nm-diameter. The diameter of diamond nanopit pattern was in good agreement with that of GC mold. Moreover, the depth of the diamond nanopit patterns fabricated by RTC-NIL using cylindrical GC mold was three times larger than that using conical diamond mold. © 2012 Materials Research Society.

关键词： Chemical vapor deposition

来源：评论

学校读者我要写书评

暂无评论

Crack Detection Based on Gaussian Mixture Model using Image Filtering

Crack Detection Based on Gaussian Mixture Model using Image ...

引用

International Symposium on Electrical and Electronics engineering (ISEE)

作者： Shujiro Ogawa Kousuke Matsushima Osamu Takahashi Advanced Engineering Course National Institute of Technology Kurume College Fukuoka Japan Department of Control and Information Systems Engineering National Institute of Technology Kurume College Fukuoka Japan Department of Civil and Environmental Engineering Nagaoka University of Technology Niigata Japan

Pavement cracks are caused by various factors such as aged deterioration, load and weather conditions, and so on. As these reduce the safety of road traffic, regular inspections are necessary. In recent years, various crack detection methods using pavement images have been proposed. However, they often have problems with accuracy and processing time. Therefore, we considered the crack detection from the viewpoint of image segmentation. In this paper, we propose a new crack detection method combining GMM and image processing which is filtering. The experimental results show that our proposed method is superior to the state-of-the-art crack detection methods in both accuracy and processing time.

关键词： Roads Image segmentation Rough sets Sparse matrices Matrix converters Gaussian mixture model

来源：评论

学校读者我要写书评

暂无评论

Fabrication of DLC-based micro-gear patterns by room-temperature curing nanoimprint lithography using glass-like carbon molds

Fabrication of DLC-based micro-gear patterns by room-tempera...

引用

2012 MRS Fall Meeting

作者： Kiyohara, Shuji Ikegaki, Tomu Ito, Chigaya Ishikawa, Ippei Tanoue, Hideto Takikawa, Hirofumi Taguchi, Yoshio Sugiyama, Yoshinari Omata, Yukiko Kurashima, Yuichi Electric and Control System Engineering Course Faculty of Advanced Engineering Maizuru National College of Technology 234 Aza Shiroya Maizuru Kyoto 625-8511 Japan Department of Electrical and Electronic Information Engineering Toyohashi University of Technology 1-1 Hibarigaoka Tenpaku Toyohashi Aichi 441-8580 Japan Application and Technical Section ELIONIX INC 3-7-6 Motoyokoyama Hachioji Tokyo 192-0063 Japan Research Center for Ubiquitous MEMS and Micro Engineering AIST 1-2-1 Namiki Tsukuba Ibaraki 305-8564 Japan

ISBN: (纸本)9781632661043

The fabrication of diamond-like carbon (DLC) micro-gear by room temperature curing nanoimprint lithography (RTC-NIL) using glass-like carbon (GC) molds as applications to the DLC-based medical MEMS (Micro Electronic Mechanical Systems) was investigated. The DLC film which has excellent properties similar to chemical vapor deposited (CVD) diamond films was used as the patterning material. We propose GC as mold material because GC has higher etching selectivity than a diamond film. The etching selectivity of polysiloxane film against a GC substrate is about 5 times as high as that of a diamond film. Therefore we fabricated the GC molds that have micro-gear patterns with 30 μm-tip diameter and 500 nm-tooth thickness. We carried out the RTC-NIL process using the GC micro-gear molds under the following optimum conditions. 1 min-time from spin-coating to imprint: t1, 0.5 MPa-imprinting pressure: P and 5 min-holding time: t2, and then the imprinted polysiloxane pattern on DLC film was processed with an electron cyclotron resonance (ECR) oxygen ion shower. However, we were not able to fabricate micro-gear patterns in high accuracy because of a remaining residual layer on the DLC film. Therefore we propose the removing process for the residual layer with trifluoromethane (CHF3) ion shower under the optimum conditions of 300 eV-ion energy and 4 min-etching time. As a result, we succeeded to fabricate concave DLC-based micro-gear patterns in high accuracy which has 30 μm-tip diameter and 1 μm-depth. © 2013 Materials Research Society.

关键词： Molds

来源：评论

学校读者我要写书评

暂无评论

Fabrication of micro-OLEDs by room-temperature curing nanocontact-print lithography using DLC molds

Fabrication of micro-OLEDs by room-temperature curing nanoco...

引用

2012 MRS Fall Meeting

作者： Ishikawa, Ippei Sakurai, Keisuke Kiyohara, Shuji Okuno, Taisuke Tanoue, Hideto Suda, Yoshiyuki Takikawa, Hirofumi Taguchi, Yoshio Sugiyama, Yoshinari Omata, Yukiko Kurashima, Yuichi Electric and Control System Engineering Course Faculty of Advanced Engineering Maizuru National College of Technology 234 Shiroya Maizuru Kyoto 625-8511 Japan Department of Electrical and Electronic Information Engineering Toyohashi University of Technology 1-1 Hibarigaoka Tenpaku Toyohashi Aichi 441-8580 Japan Application and Technical Section Elionix Inc. 3-7-6 Motoyokoyama Hachioji Tokyo 192-0063 Japan Research Center for Ubiquitous MEMS and Micro Engineering AIST 1-2-1 Namiki Tsukuba Ibaraki 305-8564 Japan

ISBN: (纸本)9781632661043

The microfabrication technologies for organic light-emitting devices (OLEDs) are essential to the fabrication of the next generation of light-emitting devices. The micro-OLEDs fabricated by room-temperature curing nanoimprint lithography (RTC-NIL) using diamond molds have been investigated. However, light emissions from 10 μm-square-dot OLEDs fabricated by the RTC-NIL method have not been uniform. Therefore, we proposed the fabrication of micro-OLEDs by room-temperature curing nanocontact-print lithography (RTC-NCL) using the diamond-like carbon (DLC) mold. The DLC molds used in RTC-NCL were fabricated by an electron cyclotron resonance (ECR) oxygen ion shower with polysiloxane oxide mask in electron beam (EB) lithography technology. The mold patterns are square and rectangle dots which has 10 μm-width, 10 μm-width and 50 μm-length, respectively. The height of the patterns is 500 nm. The DLC molds were used to form the insulating layer of polysiloxane in RTC-NCL. We carried out the RTC-NCL process using the DLC mold under the following optimum conditions: 0.1 MPa-pressure for coating DLC mold with polysiloxane film, 2.1 MPa-pressure for transferring polysiloxane from DLC mold pattern to indium tin oxide (ITO) glass substrate. We deposited N,N′-Diphenyl -N,N′-di (m-tolyl)benzidine (TPD) [40 nm-thickness] as hole transport layer / Tris(8-quinolinolato)aluminum (Alq3) [40 nm-thickness] as electron transport layer / A1 [200 nm-thickness] as cathode on ITO glass substrate as anode in this order. We succeeded in formation of the insulating layer with square and rectangle dots which has 10 μm-width, 10 μm-width and 50 μm-length, and operation of micro-OLEDs by RTC-NIL using DLC molds. © 2012 Materials Research Society.

关键词： Silicones

来源：评论

学校读者我要写书评

暂无评论

Diamond nanopit arrays fabricated by room-temperature nanoimprinting using diamond molds

Diamond nanopit arrays fabricated by room-temperature nanoim...

引用

2010 MRS Fall Meeting

作者： Kiyohara, Shuji Kumagai, Masaya Taguchi, Yoshio Sugiyama, Yoshinari Omata, Yukiko Kurashima, Yuichi Takikawa, Hirofumi Advanced Faculty of Electric and Control System Engineering Course Maizuru National College of Technology 234 Aza Shiroya Maizuru Kyoto 625-8511 Japan Application and Technical Section ELIONIX Inc. 3-7-6 Motoyokoyama Hachioji Tokyo 192-0063 Japan Department of Mechanical System Engineering University of Yamanashi 4-3-11 Takeda Kofu Yamanashi 400-8511 Japan Department of Electrical and Electronic Information Engineering Toyohashi University of Technology Toyohashi Aichi 441-8580 Japan

ISBN: (纸本)9781605112596

We have investigated the nanopatterning of chemical vapor deposited (CVD) diamond films in room-temperature nanoimprint lithography (RT-NIL), using a diamond nanodot mold. We have proposed the use of polysiloxane as an electron beam (EB) mask and RT-imprint resist materials. The diamond molds of cylinder dot using the RT-NIL process were fabricated with polysiloxane oxide mask in EB lithography technology. The dot in minimum diameter is 500 nm. The pitch between the dots is 2 μm, and dot has a height of about 600 nm. It was found that the optimum imprinting conditions for the RT-NIL : time from spin-coating to imprinting t1 of 1 min , pressure time t2 of 5 min, imprinting pressure P of 0.5 MPa. The imprint depth obtained after the press under their conditions was 500 nm. We carried out the RT-NIL process for the fabrication of diamond nanopit arrays, using the diamond nanodot molds that we developed. The resulting diamond nanopit arrays with 500 nm-diameter and 200 nm-depth after the electron cyclotron resonance (ECR) oxygen ion beam etching were fabricated. The diameter of diamond nanopit arrays was in good agreement with that of the diamond nanodot mold. © 2011 Materials Research Society.

关键词： Silicones

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：