Student models and artificial intelligence

This paper summarizes the role of student models in computer-assisted learning and lists some difficulties preventing their widespread use in practical teaching systems. Using a simple subtraction model as an illustration, the paper describes problems associated with the representation, content, creation, change, growth, execution, comparison, use for planning, use for monitoring and efficiency of student models. Similarities with some long-standing problems in artificial intelligence are pointed out. Despite the difficulties described, the paper concludes that work on improved student models should continue.     

An intelligent computer-assistant for stylistic instruction

This article describes an intelligent computer-assisted language instruction system that is designed to teach principles of syntactic style to students of English. Unlike conventional style checkers, the system performs a complete syntactic analysis of its input, and takes the student's stylistic intent into account when providing a diagnosis. Named STASEL for Stylistic Treatment At the Sentence Level, the system is specifically developed for the teaching of style, and makes use of artificial intelligence techniques in natural language processing to analyze free-form input sentences interactively.An important contribution of STASEL is its ability to provide stylistic guidance according to the specific writing goals of clarity and conciseness. In an attempt to remedy some of the deficiencies of existing instructional software, STASEL's design demonstrates how stylistic instruction can be effectively computerized, while laying the groundwork for the creation of intelligent tutoring systems for teaching writing.Graeme Hirst is an associate professor of computer science at the University of Toronto. He is the author of Semantic Interpretation and the Resolution of Ambiguity (Cambridge, 1987), and many papers on natural language understanding.Julie Payette received her B.Eng. in electrical engineering from McGill University (1986) and her M.A.Sc. from the University of Toronto (1990). She has recently been selected as a Canadian astronaut, and is currently training at the Canadian Space Agency in Ottawa.     

多媒体教学中的教学方法

分析了高校多媒体教学中存在的问题,认为运用合适的教学方法是改善多媒体教学效果的关键。为此,幻灯片内容应反映教学设计,体现教学方法的预先安排,并在讲解中综合运用其他教学法。     

Adaptive features of a CAL system based on information retrieval

A CAL system named STUDIO is presented. The system is under development and is being implemented on a Burroughs B1700 computer using the DMSII data management system.STUDIO is an information retrieval system provided with a teaching strategy. The instructional material to be presented is divided into frames, which are regarded as retrievable documents. The frames are classified with weighted index terms describing the contents of the frame. The index terms used within a course form a course thesaurus. The thesaurus is used when designing the structure of the course and in each individual run-time instructional decision in the course.After giving a general presentation of some methods of information retrieval, this paper shows how these methods are utilised in the STUDIO system, especially in the individualisation of instruction. Some aspects of student modelling are also discussed.     

Interactive 3D visualization of optical phenomena

Each year at colleges and universities nationwide, some 10,000 students take a course on optics, typically through a physics department. The method of instruction has changed very little in the past 40 years, and many of the textbook illustrations have remained essentially the same during this time. In particular, the illustrations rely heavily on line drawings and 2D graphs. Certain key concepts in optics possess fundamentally 3D aspects, which instructors typically demonstrate in a classroom laboratory. Currently, several computer-assisted instructional modules support teaching optics by visualizing optical phenomena. We created interactive 3D graphical modules for visualizing optical phenomena by applying 3D visualization techniques to instructional courseware     

Problem-solving models and search strategies for pattern recognition

Noting the major limitations of multivariate statistical classification and syntactic pattern recognition models, this paper presents an overview of some recent work using alternate representations for multistage and nearest neighbor multiclass classification, and for structural analysis and feature extraction. These alternate representations are based on generalizations of state-space and AND/OR graph models and search strategies developed in artificial intelligence (AI). The paper also briefly touches on other current interactions and differences between artificial intelligence and pattern recognition.     

Cai in geology

Computer-assisted instruction (CAI) is being used successfully, albeit limited, in several fields of geology. CAI allows increased efficiency in teaching and provides for increased individual instruction to the student. In general, students enjoy and benefit from the approach which should be employed as a compliment to normal teaching methods. Several sophisticated automated instructional systems are available presently and will have increasing application as geology students are introduced to quantitative methods and computers earlier in their education.     

Computation in the philosophy curriculum

Conclusion Computer-assisted instruction in philosophy has been successful in both supplementary and primary teaching applications. The programs used in computer-assisted instruction in philosophy are sophisticated, have proved to be acceptable over a long term, and embody aspects of fundamental research and knowledge in philosophy. The computer-assisted instruction effort in college-level philosophy has not been, nor is it likely to be, dominated by the simpler test-and-branch methods made possible by courseware authoring systems. The efforts in philosophy described in this paper provide a sound basis for forming expectations of how computers will affect college teaching in the future.     

Application of teaching and learning principles to computer-aided instruction

Many authors of software providing computer-aided instruction (CAI) have no background in educational theories. This paper reviews teaching theories and ways in which instructional principles can be applied to CAI. While the lecture form of teaching has often been maligned, there are techniques for enhancing lectures that can also apply to CAI. Effective communication methods, both verbal and audiovisual, are as important in computer modules as they are in face-to-face teaching. The quality of interactive questioning and the nature and timing of feedback are critical to the success of instruction by computer. Appropriate feedback can improve retention, as can the use of proper distractors in multiple-choice questions. We present steps for developing a CAI learning module based on established principles of teaching and learning, as well as proper identification of educational goals.     

Effectiveness of cognitive-load based adaptive instruction in genetics education

Research addressing the issue of instructional control in computer-assisted instruction has revealed mixed results. Prior knowledge level seems to play a mediating role in the student’s ability to effectively use given instructional control. This study examined the effects of three types of instructional control (non-adaptive program control, learner control, adaptive program control) and prior knowledge (high school, 1st year and 2nd year college students) on effectiveness and efficiency of learning in a genetics training program. The results revealed that adaptive program control led to highest training performance but not to superior post-test or far-transfer performance. Furthermore, adaptive program control proved to be more efficient in terms of learning outcomes of the test phase than the other two instructional control types. College students outperformed the high school students on all aspects of the study thereby strengthening the importance of prior knowledge in learning effectiveness and efficiency. Lastly, the interaction effects showed that for each prior knowledge level different levels of support were beneficial to learning.     

胃镜下早期胃癌计算机辅助分析研究综述

胃癌是全世界癌症死亡的第三大主要原因,胃癌的早期检测会对胃癌患者的后期治疗起到至关重要的作用。随着人工智能的发展,可以利用计算机视觉领域的机器学习模型辅助检测早期胃癌,有研究发现一些计算机辅助诊断模型的筛查率接近甚至高于医生。利用计算机辅助诊断可以及早发现胃癌以减少胃癌患者的后期治疗成本。报告了基于机器学习在胃镜下早期胃癌辅助诊断的研究现状,介绍了胃镜下早期胃癌的临床诊断方式,并基于此提出了计算机辅助诊断该疾病的技术路线,分析了不同诊断技术路线的研究特点,为计算机辅助诊断早期胃癌提供不同的切入点。总结了用于早期胃癌检测的机器学习、深度学习、目标检测模型,讨论了其应用于计算机辅助诊断的问题及挑战。     

Can CAL link the theory and practice of instruction?

Typically instructional decisions reflect experience more than theory, and learning theory falls short of successful practice. Successful teaching is dynamic. It requires instructional planning skill and ability to adapt to changes in student's capability. Good teachers monitor effects of instructional decisions on students and modify their lesson plans to improve instruction. Such feedback-controlled instructional design can link theory and practice but is a difficult skill to acquire without classroom trial-and-error.EDSIM is a CAL facility which provides practice in feedback-controlled instructional planning. It also may be used for research on instructional design. EDSIM can compress considerable teaching experience into a short period so that users can acquire the requisite flexibility in lesson planning. EDSIM simulates a class of 30 students. The users, as teacher, plans lessons which are submitted to the computer model. It updates submodels for each student and presents educational assessments after each simulated lesson. Users develop competence in instructional planning and an appreciation of CAL. Researchers can study effects of different factors.     

Intelligent tutoring systems survey

Simple computer assisted instruction (CAI) systems suffer from the fact that in general they do not know the subject matter they are teaching. Intelligent tutoring systems (ITS) use artificial intelligence (AI) formalisms to represent knowledge in order to improve on CAI systems. We survey a number of systems developed, and the emerging architecture for ITS development.     

人工智能对抗环境下的模型鲁棒性研究综述

近年来人工智能研究与应用发展迅速,机器学习模型大量应用在现实的场景中,人工智能模型的安全鲁棒性分析与评估问题已经开始引起人们的关注。最近的研究发现,对于没有经过防御设计的模型,攻击者通过给样本添加微小的人眼不可察觉的扰动,可以轻易的使模型产生误判,从而导致严重的安全性问题,这就是人工智能模型的对抗样本。对抗样本已经成为人工智能安全研究的一个热门领域,各种新的攻击方法,防御方法和模型鲁棒性研究层出不穷,然而至今尚未有一个完备统一的模型鲁棒性的度量评价标准,所以本文总结了现阶段在人工智能对抗环境下的模型鲁棒性研究,论述了当前主流的模型鲁棒性的研究方法,从一个比较全面的视角探讨了对抗环境下的模型鲁棒性这一研究方向的进展,并且提出了一些未来的研究方向。     

自然语言预训练模型知识增强方法综述

将知识引入到依靠数据驱动的人工智能模型中是实现人机混合智能的一种重要途径.当前以BERT为代表的预训练模型在自然语言处理领域取得了显著的成功,但是由于预训练模型大多是在大规模非结构化的语料数据上训练出来的,因此可以通过引入外部知识在一定程度上弥补其在确定性和可解释性上的缺陷.该文针对预训练词嵌入和预训练上下文编码器两个...     

Relevance vector machines approach for long-term flow prediction

Over the past years, some artificial intelligence techniques like artificial neural networks have been widely used in the hydrological modeling studies. In spite of their some advantages, these techniques have some drawbacks including possibility of getting trapped in local minima, overtraining and subjectivity in the determining of model parameters. In the last few years, a new alternative kernel-based technique called a support vector machines (SVM) has been found to be popular in modeling studies due to its advantages over popular artificial intelligence techniques. In addition, the relevance vector machines (RVM) approach has been proposed to recast the main ideas behind SVM in a Bayesian context. The main purpose of this study is to examine the applicability and capability of the RVM on long-term flow prediction and to compare its performance with feed forward neural networks, SVM, and multiple linear regression models. Meteorological data (rainfall and temperature) and lagged data of rainfall were used in modeling application. Some mostly used statistical performance evaluation measures were considered to evaluate models. According to evaluations, RVM method provided an improvement in model performance as compared to other employed methods. In addition, it is an alternative way to popular soft computing methods for long-term flow prediction providing at least comparable efficiency.     

基于C/S架构的教学系统中智能技术的应用研究

利用人工智能原理及认知理论建立ICAI模型,构建实用的教学系统是人工智能的一个重要应用.本文就ICAI的特点、教学系统的整体结构、功能模块及其关键技术进行了详细的分析,并对教学过程进行了定量分析.     

Using background knowledge in case-based legal reasoning: A computational model and an intelligent learning environment

Researchers in the field of AI and Law have developed a number of computational models of the arguments that skilled attorneys make based on past cases. However, these models have not accounted for the ways that attorneys use middle-level normative background knowledge (1) to organize multi-case arguments, (2) to reason about the significance of differences between cases, and (3) to assess the relevance of precedent cases to a given problem situation. We present a novel model, that accounts for these argumentation phenomena. An evaluation study showed that arguments about the significance of distinctions based on this model help predict the outcome of cases in the area of trade secrets law, confirming the quality of these arguments. The model forms the basis of an intelligent learning environment called CATO, which was designed to help beginning law students acquire basic argumentation skills. CATO uses the model for a number of purposes, including the dynamic generation of argumentation examples. In a second evaluation study, carried out in the context of an actual legal writing course, we compared instruction with CATO against the best traditional legal writing instruction. The results indicate that CATO's example-based instructional approach is effective in teaching basic argumentation skills. However, a more “integrated” approach appears to be needed if students are to achieve better transfer of these skills to more complex contexts. CATO's argumentation model and instructional environment are a contribution to the research fields of AI and Law, Case-Based Reasoning, and AI and Education.