A knowledge extraction and representation system for narrative analysis in the construction industry

Many researchers advocate that the real-world narratives shared by experts or knowledge workers are helpful in teaching and educating novices to learn new knowledge and skills. Narrative analysis is a useful method for experts to understand narratives. However, it does not produce any clear or explicit layouts. This is not easy for a new learner without prior knowledge to glean the right messages from narratives within a short time. In this paper, a narrative knowledge extraction and representation system (NKERS) is presented to extract and represent narrative knowledge in an effective manner. The NKERS is composed of a narrative knowledge element extraction algorithm, a narrative knowledge representation method and a narrative knowledge database. A prototype system has been built and trial implemented in the construction industry. The results show that the domain experts agree that the narrative maps generated by the NKERS can effectively represent narrative elements and flows. Three-quarters of respondents expressed that they will use the produced narrative maps in their training courses to facilitate students’ learning.     

A narrative-based reasoning with applications in decision support for social service organizations

Nowadays, there is an increasing demand for incorporating unstructured narratives in decision support for knowledge-intensive industries such as healthcare and social service organizations. However, most of the current research on decision support systems (DSS) mainly focused on dealing with structured data and are inadequate to dealing with unstructured narratives such as clients’ records and stories. This paper presents a narrative-based reasoning (NBR) algorithm which incorporates the technologies of knowledge-based system (KBS), computational linguistics, and artificial intelligence (AI) for automatic processing unstructured narratives and inferring useful knowledge for decision support. A NBR enabled DSS has been built and was evaluated through a series of experiments conducted in early intervention of mental health of a social service company in Hong Kong. The performance of NBR was measured based on recall and precision and encouraging results were obtained. High recall and precision are achieved in the reasoning of unstructured data, and high recall is achieved for the association analysis. The results show that it is possible for inferring recommendations for problem solving from unstructured narratives automatically. Based on the approach, it helps to support knowledge workers with reliable suggestions on decision making so as to increase the quality of their solutions.     

An empirical study of narrative imagery in implicit and explicit contexts

This study investigates if imagery is an inherent construct to narrative by playing a role in storytelling and comprehension. Visualization activities consist of ones that depend on pictures or physical objects and those do not depend on extraneous visual artifacts. The understandings of both types of visualization will contribute to their application and integration. In light of rapid development of technology and drastic increase of multimedia representations in social communication, applying storytelling approach to system design is increasingly relevant to many researchers who are eager to bring visual thinking to the classroom, alternative to applying physical visual artifacts. In this study, we looked into the cognitive process evoked in the comprehension of narrative and its similarity to imagery as an individual cognition. We transformed a course into conversational narrative and participants were randomly assigned into three sections, two were in narrative text and one was in expository text. Sections 1 and 2 were in narrative text, but only participants from section 1 were prompt for image creation before writing essays. Section 3 was in expository text and participants were also prompt for image creation before writing essays. The independent samples t-test was used to compare the mean scores of three groups on creativity score and word total across the sections. Our study found that narrative was able to activate the imagery world of the participants without giving them further instructions explicitly suggesting so. The implicit imagery context created by reading narratives had a stronger impact on creativity than that of explicit imagery creation context that did not involve of narrative comprehension. The study suggests that narrative innately provides some sort of control to its user cognitively and can be integrated with other types of media in design.     

Narrative prose generation

Narrative generation has historically suffered from poor writing quality, stemming from a narrow focus on story grammars and plot design. Moreover, to-date natural language generation systems have not been capable of faithfully reproducing either the variety or complexity of naturally occurring narratives. In this article we first propose a model of narrative derived from work in narratology and grounded in observed linguistic phenomena. Next we describe the Author architecture for narrative generation and an end-to-end implementation of the Author model in the StoryBook narrative prose generation system. Finally, we present a formal evaluation of the narratives that StoryBook produces.     

The nature of research: qualitative or quantitative,narrative or paradigmatic?

Abstract. This paper explores the role of qualitative and quantitative approaches in information systems research. The importance of qualitative data and the construction of narratives as a key procedure in the development of theoretical conjectures and empirical generalizations or hypotheses are discussed. This paper also puts the importance of paradigmatic thinking and research into perspective by contrasting this approach with narrative thinking and qualitative research.
A recent PhD dissertation on strategic information systems is used to illustrate some of the concepts discussed in the paper.
Although the paper advocates the importance of qualitative research and argues that in some respects it may be regarded as more creative than quantitative research, it clearly recognizes that the distinguishing feature of modern science is the formulation of laws, very often mathematical, that capture the essential features of a problem and allow us to make predictions both into the future and about other current situations.     

How causal knowledge simplifies decision-making

Making decisions can be hard, but it can also be facilitated. Simple heuristics are fast and frugal but nevertheless fairly accurate decision rules that people can use to compensate for their limitations in computational capacity, time, and knowledge when they make decisions [Gigerenzer, G., Todd, P. M., & the ABC Research Group (1999). Simple Heuristics That Make Us Smart. New York: Oxford University Press.]. These heuristics are effective to the extent that they can exploit the structure of information in the environment in which they operate. Specifically, they require knowledge about the predictive value of probabilistic cues. However, it is often difficult to keep track of all the available cues in the environment and how they relate to any relevant criterion. This problem becomes even more critical if compound cues are considered. We submit that knowledge about the causal structure of the environment helps decision makers focus on a manageable subset of cues, thus effectively reducing the potential computational complexity inherent in even relatively simple decision-making tasks. We review experimental evidence that tested this hypothesis and report the results of a simulation study. We conclude that causal knowledge can act as a meta-cue for identifying highly valid cues, either individual or compound, and helps in the estimation of their validities.     

计算受限控制系统的一种全资源预测控制方法

针对具有时变有限且不可预知计算资源的控制系统，提出了一种充分利用可用计算资源的预测控制策略和相应的控制器设计方法.该策略在控制系统可用计算资源充足时计算多步前向预测控制量，进而使用合适预测控制量在控制器因缺少计算资源无法运行时闭合系统，达到了在不要求额外计算资源前提下提升控制系统性能的效果.利用改进的模型预测控制方法设计了相应的控制器，并分别使用纯数值和MATLAB/LabVIEW联合仿真算例对所提出的方法进行了验证.     

Process knowledge-based random forest regression for model predictive control on a nonlinear production process with multiple working conditions

In process industry, predictive control approaches have been widely used for nonlinear production processes. Practically, the predictor in a predictive controller is extremely important since it provides future states for the optimization problem of controllers. The conventional predictive controller with precise mathematical predictors approximating the state space of physical systems is difficult and time-consuming for nonlinear production processes, and it performs poorly over a wide range of working conditions and with significant disturbances. To address the challenges, the trend of applying artificial intelligence emerges. However, the industrial process-specific knowledge is ignored in most cases. In this study, a predictive controller with a control process knowledge-based random forest (RF) model is proposed. Specifically, working data are clustered at first to handle diverse working conditions. Then, a process knowledge-based forest predictor, namely MIW-RF model with a redesigned cascading RF structure, is proposed to incorporate control process knowledge into modeling. Thus, future states of controlled variables could be more accurately acquired for the optimizer. A simplified version of the predictive model is also developed with quick model training and updating. The proposed predictive methods are finally introduced into the controller design. According to the empirical results, the proposed methods deliver a better control performance against benchmarks, including more accurate anticipated controlled-variable responses, better set-point tracking and disturbance rejection capability.     

A new solution to optimization-satisfaction problems by a penalty method

This paper is concerned with an optimization-satisfaction problem to determine an optimal solution such that a certain objective function is minimized, subject to satisfaction conditions against uncertainties of any disturbances or opponents' decisions. Such satisfaction conditions require that plural performance criteria are always less than specified values against any disturbances or opponents' decisions. Therefore, this problem is formulated as a minimization problem with the constraints which include max operations with respect to the disturbances or the opponents' decision variables. A new computational method is proposed in which a series of approximate problems transformed by applying a penalty function method to the max operations within the satisfaction conditions are solved by usual nonlinear programming. It is proved that a sequence of approximated solutions converges to a true optimal solution. The proposed algorithm may be useful for systems design under unknown parameters, process control under uncertainties, general approximation theory, and strategic weapons allocation problems.     

Effective optimization for fuzzy model predictive control

This paper addresses the optimization in fuzzy model predictive control. When the prediction model is a nonlinear fuzzy model, nonconvex, time-consuming optimization is necessary, with no guarantee of finding an optimal solution. A possible way around this problem is to linearize the fuzzy model at the current operating point and use linear predictive control (i.e., quadratic programming). For long-range predictive control, however, the influence of the linearization error may significantly deteriorate the performance. In our approach, this is remedied by linearizing the fuzzy model along the predicted input and output trajectories. One can further improve the model prediction by iteratively applying the optimized control sequence to the fuzzy model and linearizing along the so obtained simulated trajectories. Four different methods for the construction of the optimization problem are proposed, making difference between the cases when a single linear model or a set of linear models are used. By choosing an appropriate method, the user can achieve a desired tradeoff between the control performance and the computational load. The proposed techniques have been tested and evaluated using two simulated industrial benchmarks: pH control in a continuous stirred tank reactor and a high-purity distillation column.     

An ontology-driven decision support system for high-performance and cost-optimized design of complex railway portal frames

Electrification structures design for railway systems is a crucial and complex process, since it compounds plenty of infrastructure elements, design decisions, and calculation conditions. In this paper, an ontology-driven decision support system for designing complex railway portal frames is presented and developed. A knowledge-rules database has been also developed relying on experts knowledge and complying with railway standards. Our system outperforms the current portal frames design methods by decreasing construction time and costs. As a result, an intelligent computer-aided design tool is provided, thus facilitating the task of seeking for the optimal portal frame, which is geometrically and structurally feasible, and cost-effective.     

基于广义投影神经网络优化的模型预测控制*

为降低模型预测控制优化问题的计算复杂度,以时滞系统的模型预测控制问题作为研究对象,利用神经网络动态平衡点与优化问题解相对应的特点,提出一种基于广义投影神经网络的模型预测控制优化算法。首先,将模型预测控制优化问题描述为一个带约束的二次规划问题,进一步,通过广义投影神经网络模型进行在线优化。该方法充分发挥了神经网络并行、结构简单的优点,通过具体实例仿真,验证了本文算法的有效性与优越性。     

Unsupervised learning of generative and discriminative weights encoding elementary image components in a predictive coding model of cortical function

A method is presented for learning the reciprocal feedforward and feedback connections required by the predictive coding model of cortical function. When this method is used, feedforward and feedback connections are learned simultaneously and independently in a biologically plausible manner. The performance of the proposed algorithm is evaluated by applying it to learning the elementary components of artificial and natural images. For artificial images, the bars problem is employed, and the proposed algorithm is shown to produce state-of-the-art performance on this task. For natural images, components resembling Gabor functions are learned in the first processing stage, and neurons responsive to corners are learned in the second processing stage. The properties of these learned representations are in good agreement with neurophysiological data from V1 and V2. The proposed algorithm demonstrates for the first time that a single computational theory can explain the formation of cortical RFs and also the response properties of cortical neurons once those RFs have been learned.     

Problem-oriented simulations to develop and improve higher-order thinking strategies

The purpose of this article is to present an empirically based instructional method to improve higher-order thinking strategies. The method employs computer-managed simulations that present contextually meaningful problem situations that require students to prepare solution proposals. The simulation assesses the proposal and offers the students the consequences of their decisions while also iteratively updating the situational conditions. This type of simulation, unlike conventional simulations which are used for acquisition of knowledge, is problem-oriented, requiring the students to fully employ their knowledge base by generating solutions to domain-specific problems. Thus, the focus of problem-oriented simulations is to improve student cognitive abilities employed in the service of recall, problem solving, and creativity.     

一种基于连通分量的文本区域定位方法

文本区域定位对复杂背景图像中的字符识别和检索具有重要意义。已有方法取得高的定位准确率和召回率,但效率较低,难以应用于实际的系统中。文中提出一种基于连通分量过滤和K-means聚类的文本区域定位方法。该方法首先对图像进行自适应分割,对字符颜色层提取连通分量。然后提取连通分量的特征,并用Adaboost分类器过滤非字符连通分量。最后,对候选的字符连通分量根据其位置和颜色层进行K-means聚类来定位文本区域。实验结果显示该方法具有与当前方法相当的准确率和召回率,同时具有较低的计算复杂度。     

Towards Ontology-driven Knowledge Synthesis for Heterogeneous Information Systems

Information integration enables delivery of the right information to the right user in a timely manner giving manufacturers a competitive edge in today’s global manufacturing market. However, as enterprise information is usually aggregated from a variety of heterogeneous information sources, without using an adequate integration framework it is difficult to extract pertinent information and apply current knowledge to assessing production situations and making informed decisions. This paper investigates a method of facilitating knowledge synthesis in a distributed computing environment. A formal model of domain ontology and knowledge base is presented, which aims at providing a vehicle for representing information and knowledge using a common shared semantics in a given application domain. As a result, a common knowledge representation based architecture is proposed, creating a foundation for establishing a systematic approach for ease of knowledge synthesis in a manufacturing environment     

A hybrid knowledge and ensemble classification approach for prediction of venous thromboembolism

Clinical narratives such as progress summaries, lab reports, surgical reports, and other narrative texts contain key biomarkers about a patient's health. Evidence-based preventive medicine needs accurate semantic and sentiment analysis to extract and classify medical features as the input to appropriate machine learning classifiers. However, the traditional approach of using single classifiers is limited by the need for dimensionality reduction techniques, statistical feature correlation, a faster learning rate, and the lack of consideration of the semantic relations among features. Hence, extracting semantic and sentiment-based features from clinical text and combining multiple classifiers to create an ensemble intelligent system overcomes many limitations and provides a more robust prediction outcome. The selection of an appropriate approach and its interparameter dependency becomes key for the success of the ensemble method. This paper proposes a hybrid knowledge and ensemble learning framework for prediction of venous thromboembolism (VTE) diagnosis consisting of the following components: a VTE ontology, semantic extraction and sentiment assessment of risk factor framework, and an ensemble classifier. Therefore, a component-based analysis approach was adopted for evaluation using a data set of 250 clinical narratives where knowledge and ensemble achieved the following results with and without semantic extraction and sentiment assessment of risk factor, respectively: a precision of 81.8% and 62.9%, a recall of 81.8% and 57.6%, an F measure of 81.8% and 53.8%, and a receiving operating characteristic of 80.1% and 58.5% in identifying cases of VTE.     

Mediating effects of active and distributed instruments on narrative activities

This paper discusses the effects of introducing new distributed and active instruments on narrative activities in a school environment. We address the issue of how the Pogo instruments change children's activity when they invent stories. The results enable us to compare the way the activity is carried out, both in its conventional context and with the Pogo instruments, mainly along three main lines of investigation: the collective dimension, the use of space and the structure of the narrative. The results also show that using the instruments increase the collective or group dimension of the creative process, particularly the role diversification and participation of the children. These instruments support children's efforts to structure narratives and thereby produce richer stories.

This research was carried out within the Pogo Project by a multidisciplinary team that included interactive design and user-centered approaches within the EC I3 programme on ‘Exploring New Learning Futures for Children’.     

Nonlinear model predictive control of multivariable processes using block-structured models

Block-structured models, such as Wiener or Hammerstein models, have been used in nonlinear model predictive control to reduce the cost of identification and online computation. The solution of a nonlinear dynamic optimization problem has been avoided by inverting the nonlinear element and solving the resulting linear problem in the past. However, by exploiting the block structure for sensitivity calculation, the original nonlinear problem can also be solved at low computational cost. At the same time, greater modeling flexibility is achieved. Recently, a new Hammerstein model structure has been proposed for multivariable processes with input directionality, which exploits such increased modeling flexibility. This paper deals with nonlinear model predictive control constrained by models of Hammerstein or Uryson structure. A method for efficient calculation of sensitivity information is developed. In a simulation example, the method is shown to combine low computational cost with a significant reduction of the loss of optimality compared to the previous methods.     

基于l0范数近似最小化的稀疏信号重构方法

针对稀疏信号的重构问题,提出了[l0]范数近似最小化算法。利用反正切函数近似[l0]范数建立相应的非凸优化问题。通过构造快速的不动点迭代格式求解该问题,分析了所提出算法的收敛性能。数值仿真表明,该算法具有重构信号需要测量值少、计算精度高且计算量较小的优点。