The training of knowledge structures for manufacturing tasks: an empirical study

Abstract

Knowledge structure refers to the manner in which a human organizes knowledge with a given domain. Research has identified knowledge structure as a determinant of the human ability to perform cognitive-oriented tasks. Yet uncertainty still exists about how to improve an individual's cognitive task performance through the controlled utilization of the individual's knowledge structure. The purpose of this study is to investigate whether the development of individual's knowledge structure in a particular domain can be manipulated through training. The experiment utilized the manufacturing domain of plastic extrusion machine operation. Sixteen subjects, having no previous knowledge of the domain, were randomly assigned to one of two experimental groups. Each of the experimental groups corresponded to a distinct training condition. Over a three-day period, both training groups received the same instructional content; however, the sequence in which the training material was presented differed. One group initially received the abstract, conceptual relationships between domain concepts, followed by more detailed relationships associated with the lower level aspects of the domain. The other group received the training material in the reverse order; i.e. the lower level information followed by the abstract. Prior to and concluding the training sessions, each individual's knowledge structure was assessed along two dimensions, hierarchical levels and multiple relations, through a computer-based measurement technique.entitled KSAT. The group which received the abstract relationships first showed significant improvement following training along both dimensions of knowledge structure. No significant changes in the knowledge structure dimensions were found for the group which received the lower level relationships first. This study suggests that an individual's knowledge structure can be manipulated through training, with a significant effect being attributed to the training sequence of abstract material followed by the more detailed material.     

通用型CAPP专家系统中知识表示和推理机制的研究

ＣＡＰＰ作为ＣＡＤ与ＣＡＭ的中间桥梁，是ＣＩＭＳ必不可少的重要环节．本文针对开发通用性ＣＡＰＰ专家系统过程中，产品零件表示和工艺规程设计这两个关键问题，根据事物之间的相似性原理，提出了面向对象的产品零件和工艺规程的知识表示方法，以满足ＣＡＰＰ系统对产品零件信息表示的完整性、通用性、逻辑可分解性、动态性和相容性等要求，并在此基础上提出了ＣＡＰＰ专家系统的混合推理机制，包括单继承、多继承、非单调继承、类比推理等．     

Introduction to information systems theory: concepts,formalism and applications

An aim of the proposed information systems theory ?IST) is to build a bridge between the general systems theory's formalism and the world of information and information technologies, dealing with transformation of information as a common non-material substance, whose models in forms of computer algorithms and programs could be implemented to different material objects, including a human's thoughts and languages. A new approach to IST is based on a single concept and follows mathematical formalism that uses an information variation principle to build an information systemic model of a specific object. The problem's solution, procedure and methodology modelling are illustrated by the application of the information macrodynamics ?IMD), which reveals the system model's main layers: microlevel stochastics, macrolevel dynamics, hierarchical dynamic network ?IN) of information structures, its minimal logic, and optimal code of communication language, generated by the IN hierarchy, dynamics and geometry. The system's complex dynamics originate information geometry and evolution with the functional information mechanisms of ordering, cooperation, mutation, stability, diversity, adaptation, self-organization, the double spiral's genetics, the system's generation, decaying, and transferring information to other systems along with the information mechanisms of heredity and replication. The developed IMD's theoretical computer-based methodology and software have been applied to such areas as technology, communications, computer science, intelligent processes, biology, economy, management, and other non-physical and physical subjects with their mutual interactions, informational superimposition, and the information transferred between interactions.     

CAUSAL BAYESIAN REASONING IN MEDICINE

Abstract

We propose a knowledge representation architecture organized in three levels—a causal network (containing the domain knowledge), medical strategies and causal reasoning—and expose the fundamentals of a mathematical model for computing probaility through the application of Bayes' theorem in a causal network.     

The impact of hierarchical positions on the type of communication within online Communities of Learning

Communities of Learning (CoL) have been suggested to facilitate the learning process among participants of online trainings. Yet, previous studies often detached participants from the social context in which learning took place. The present study addresses this shortcoming by providing empirical evidence from 25 CoL of a global organization, where 249 staff members from different hierarchical positions engaged into collaborative learning via asynchronous discussion forums. We conduct a longitudinal study on the type of communication within these CoL, as well as participants' network positions, in order to investigate the research question: What is the impact of individual's hierarchical positions on the type of communication within CoL? Our results indicate that the higher participants' hierarchical position, the higher their amount of social and cognitive communication, which in turn was also positively related to their network position within CoL. We also identified a sub-group of Stars that outperformed their colleagues and who were at the center of CoL, irrespective of their hierarchical positions. Consequently, we propose design and facilitation strategies to practitioners and organizers of future CoL that can foster the learning processes and outcomes of all participants. Additionally, we consider future research avenues that could be explored further.     

一种融合节点先验信息的图表示学习方法

图表示学习是实现各类图挖掘任务的基础。现实当中的图数据,不仅包含复杂的网络结构,还包括多样化的节点信息。如何将网络结构和节点信息更加有效地融入图的表示学习中,是一个重要的问题。为了解决这一问题,本文基于深度学习提出了融合节点先验信息的图表示学习方法。该方法将节点特征作为先验知识,要求学习到的表示向量同时保持图数据中的网络结构相似性和节点特征相似性。该方法的时间复杂度为O（|V|）,其中|V|为图节点数量,表明该方法适用于大规模图数据分析。同时,在多个数据集上的实验结果表明,所提出的方法相比目前流行的几种基线方法,在分类任务上能够获得良好而稳定的优势。     

Progressive decomposition: a method of coarse-to-fine image parsing using stacked networks

To parse images into fine-grained semantic parts, the complex elements will put it in trouble when using off-the-shelf semantic segmentation networks, because it is difficult for them to utilize the contextual information of fine-grained parts. In this paper we propose a progressive decomposition method to parse images in a coarse-to-fine manner with refined semantic classes. It consists of two aspects: stacked networks and progressive supervisions. The stacked network is achieved by stacking some segmentation layers in a segmentation network. The former segmentation module parses images at a coarser-grained level, and the result will be fed to the following one to provide effective contextual clues for the finer-grained parsing. The skip connections from shallow layers of the network to fine-grained parsing modules are also added to recover the details of small structures. For the training of the stacked networks which have coarse-to-fine outputs, a strategy of progressive supervision is proposed to merge classes in ground truth to get coarse-to-fine label maps, and then train the stacked network end-to-end with the hierarchical supervisions. The proposed framework can be injected into many advanced neural networks to improve the parsing results. Extensive evaluations on several public datasets including face parsing and human parsing well demonstrate the superiority of our method.

     

Intuition, Insight, Imagination and Creativity

Can computers have intuition and insights, and be creative? Neurocognitive models inspired by the putative processes in the brain show that these mysterious features are a consequence of information processing in complex networks. Intuition is manifested in categorization based on evaluation of similarity, when decision borders are too complex to be reduced to logical rules. It is also manifested in heuristic reasoning based on partial observations, where network activity selects only those paths that may lead to solution, excluding all bad moves. Insight results from reasoning at the higher, non-verbal level of abstraction that comes from involvement of the right hemisphere networks forming large "linguistic receptive fields." Three factors are essential for creativity in invention of novel words: knowledge of word morphology captured in network connections, imagination constrained by this knowledge, and filtering of results that selects the most interesting novel words. These principles have been implemented using a simple correlation-based algorithm for auto-associative memory. Results are surprisingly similar to those created by humans.     

THE CHALLENGE OF DEEP MODELS,INFERENCE STRUCTURES,AND ABSTRACT TASKS

Abstract

An attempt is made to see recent developments in knowledge representation, clustered around Minsky 's theory of frames, and the more classical modeling of natural processes, originating in Newton's mechanics, in a unified perspective. Defining knowledge as homomorphic representation, we view classical models as capable of providing behavioral knowledge but lacking in capability to represent important structural knowledge and meta-knowledge. He discuss a framework, inspired by system theoretic concepts, for combining structural and behavioral representation in a form suitable for goal-directed model construction. We conclude by drawing several implications of this approach for system design, artificial intelligence, modeling, and simulation.     

Viewing personal history records: A comparison of tabular format and graphical presentation using LifeLines

Thirty-six participants used a static version of either LifeLines, a graphical interface, or a tabular representation to answer questions about a database of temporal personal history information. Results suggest that overall the LifeLines representation led to much faster response times, primarily for questions which involved interval comparisons and making intercategorical connections. A 'first impression' test showed that LifeLines can reduce some of the biases of the tabular record summary. A post-experimental memory test led to significantly (p< 0.004) higher recall for LifeLines. Finally, simple interaction techniques are proposed to compensate for the problems of the static LifeLines display's ability to deal with precise dates, attribute coding and overlaps.     

Analysis of Mental Processes Involved in Air Traffic Control

Abstract

The research which is briefly presented here has been carried out, during the lost ten years, by a team of psychologists at the Centre d'Experimentation de la Navigation A?rienne, whose aim is the automation of the air traffic control system. The general topic refers to the operator-computer's interactions in a complex system. The type of contribution that psychology may provide is shown, by indicating the three interdependent aspects of the research : 1) the definition of the operator's algorithms; 2) the analysis of his ‘mental representation’; 3) the study of the effect of the work load on the mental processes.

The general methodology of the studies of mental processes is presented; and an experiment on operative memory is discussed.     

Multi-Agent Dynamic Logics with Informational Test

This paper investigates a family of logics for reasoning about the dynamic activities and informational attitudes of agents, namely the agents' beliefs and knowledge. The logics are based on a new formalisation and semantics of the test operator of propositional dynamic logic and a representation of actions which distinguishes abstract actions from concrete actions. The new test operator, called informational test, can be used to formalise the beliefs and knowledge of particular agents as dynamic modalities. This approach is consistent with the formalisation of the agents' beliefs and knowledge as K(D)45 and S5 modalities. Properties concerning informativeness, truthfulness and preservation of beliefs are proved for a derivative of the informational test operator. It is shown that common belief and common knowledge can be expressed in the considered logics. This means, the logics are more expressive than propositional dynamic logic with an extra modality for belief or knowledge. The logics remain decidable and belong to 2EXPTIME. Versions of the considered logics express natural additional properties of beliefs or knowledge and interaction of beliefs or knowledge with actions. It is shown that a simulation of PDL can be constructed in one of these extensions.     

Cross-media similarity metric learning with unified deep networks

As a highlighting research topic in the multimedia area, cross-media retrieval aims to capture the complex correlations among multiple media types. Learning better shared representation and distance metric for multimedia data is important to boost the cross-media retrieval. Motivated by the strong ability of deep neural network in feature representation and comparison functions learning, we propose the Unified Network for Cross-media Similarity Metric (UNCSM) to associate cross-media shared representation learning with distance metric in a unified framework. First, we design a two-pathway deep network pretrained with contrastive loss, and employ double triplet similarity loss for fine-tuning to learn the shared representation for each media type by modeling the relative semantic similarity. Second, the metric network is designed for effectively calculating the cross-media similarity of the shared representation, by modeling the pairwise similar and dissimilar constraints. Compared to the existing methods which mostly ignore the dissimilar constraints and only use sample distance metric as Euclidean distance separately, our UNCSM approach unifies the representation learning and distance metric to preserve the relative similarity as well as embrace more complex similarity functions for further improving the cross-media retrieval accuracy. The experimental results show that our UNCSM approach outperforms 8 state-of-the-art methods on 4 widely-used cross-media datasets.

     

INFORMATION MACRODYNAMIC MODEL AND INFORMATION STRUCTURE OF A MARKET ECONOMIC SYSTEM

Information represents a common and universal substance, active participating in a diversity of physical or virtual interactions, including various forms of economic interactions.

The math-information controlled model based on an information path functional evaluates the information regularities of macrodynamic economic processes.

The built information systemic models focus on an elementary production-organization, the production's interaction and management, and different market dynamics. An organization is modeled by the hierarchical structure of an information cooperative dynamic space distributed network.

The found formalized information mechanisms govern market's cooperative dynamics, and impose the information restrictions on these processes.

The considered systemic mechanisms of self-control, adaptation, and evolution, represent a general attribute of an economical system.     

概念表示增强的知识追踪模型

知识追踪模型以学习者的历史学习行为数据作为输入,通过概念表示来描述学习者的概念掌握状态,从而预测学习者未来的学习表现。然而在概念的外延表示方面,当前知识追踪研究的概念外延信息被限制在一阶相关的范畴内,无法表征概念的一阶以上外延信息。为了解决这一问题,提出方法首先使用图结构描述概念内涵信息及其相互关系;其次使用图神经网络的池化操作等提取概念的外延表示,这保证了概念的外延信息来源于多阶相关关系;再与概念的内涵表示进行融合;最后预测学习者未来的答题情况。为了验证该模型的有效性和效率,选取了四个主流知识追踪模型作为对比模型,在四个常用的知识追踪数据集上进行实验。结果表明,提出模型在若干评价指标上均取得了一定的优势,说明了它的有效性;在模型性能方面,提出模型达到最优评价指标所需的迭代次数最少,说明了它的效率;在实际应用方面,以该模型为基础实现了一个智能学习平台,在三门线下课程的教学过程中判断和预测学习者未来答题情况,取得了优于其他知识追踪模型的表现。     

Harnessing knowledge workers' participation for IT planning effectiveness: The informational and motivational mediating effects of users' microplanning behaviour

In most organizations, knowledge is produced on a continuous basis such that formal planning methods fail to capture it on time for utilization into effective IT solutions. Microplanning, as user-led informal planning behaviour, allows users as knowledge workers to continuously use their knowledge and skills to identify opportunities for using IT or replanning its existing use. This is the informational effect of microplanning. Microplanning is also an empowering process which enhances users' own motivation. This empowerment effect is referred to as the motivational effect. User microplanning behaviour therefore constitutes the main construct of the model of microplanning effectiveness described in this study. It yields effectiveness through two paths, informational and motivational. After operational definitions for the variables were derived, the two main hypotheses of the study were empirically tested on a sample of 263 knowledge workers. Results of the study fully support the informational and motivational effects of microplanning.     

Infoweb: An adaptive information filtering system for the cultural heritage domain

This paper presents a system developed for adaptive retrieval and the filtering of documents belonging to digital libraries available on the Web. This system, called InfoWeb, is currently in operation on the ENEA (National Entity for Alternative Energy) digital library Web site reserved to the cultural heritage and environment domain. InfoWeb records the user information needs in a user model, created through a representation, which extends the traditional vector space model and takes the form of a semantic network consisting of co-occurrences between index terms. The initial user model is built on the basis of stereotypes, developed through a clustering of the collection by using specific documents as a starting point. The user's query can be expanded in an adaptive way, using the user model formulated by the user himself. The system has been tested on the entire collection comprising about 14,000 documents in HTML/text format. The results of the experiments are satisfactory both in terms of performance and in terms of the system's ability to adapt itself to the user's shifting interests.     

图网络层级信息挖掘分类算法综述

深度学习作为人工智能的一个研究分支发展迅速,而研究数据主要是语音、图像和视频等,这些具有规则结构的数据通常在欧氏空间中表示。然而许多学习任务需要处理的数据是从非欧氏空间中生成,这些数据特征和其关系结构可以用图来定义。图卷积神经网络通过将卷积定理应用于图,完成节点之间的信息传播与聚合,成为建模图数据一种有效的方法。尽管图卷积神经网络取得了巨大成功,但针对图任务中的节点分类问题,由于深层图结构优化的特有难点——过平滑现象,现有的多数模型都只有两三层的浅层模型架构。在理论上,图卷积神经网络的深层结构可以获得更多节点表征信息,因此针对其层级信息进行研究,将层级结构算法迁移到图数据分析的核心在于图层级卷积算子构建和图层级间信息融合。本文对图网络层级信息挖掘算法进行综述,介绍图神经网络的发展背景、存在问题以及图卷积神经网络层级结构算法的发展,根据不同图卷积层级信息处理将现有算法分为正则化方法和架构调整方法。正则化方法通过重新构建图卷积算子更好地聚合邻域信息,而架构调整方法则融合层级信息丰富节点表征。图卷积神经网络层级特性实验表明,图结构中存在层级特性节点,现有图层级信息挖掘算法仍未对层级特性节点的...     

Class-Entropy minimisation networks for domain analysis and rule extraction

When applied to supervised classification problems, neural rule extraction aims at making classification mechanisms explicit by reversing the knowledge embedded in a network's connections. To this end, the present research has borrowed notions from information theory; using Conditional Class Entropy as a network cost function improves representation efficiency by forcing the network to emphasize only task-essential information. We present a library of methods to analyse, simplify and rearrange the knowledge embedded in CCE-trained networks; the final result is a hierarchy of if-then rules modelling the classification process in symbolic form. Experimental results on a clinical testbed (diagnosis of Lyme disease) confirmed the effectiveness and the generalisation power of the methodologies described. In addition, satisfactory results obtained on this still unsolved clinical problem endowed this research with interdisciplinary value.