Incremental Knowledge Acquisition Research Articles

Factors Determining Knowledge Acquisition in Livestock Waste Management Training Held by Animal Husbandry Training Centre at Songgoriti, Batu - Indonesia

Abstract The livestock industry plays a substantial role in the Indonesian economy as the industry provides a significant share of the country’s national income. The industry also engages labour in rural and sub-urban areas since traditional smallholder livestock dominates the industry in the country. One problem in the smallholder livestock is the inadequate capacity of farmers leading to mediocre livestock management, generating polluting waste and low income. A special training curriculum for smallholder farmers is expected to improve farmer capacity. Animal Husbandry Training Centre at Songgoriti regularly holds special training in waste management for smallholder livestock farmers. This study aimed to analyse the training outcome and factors influencing knowledge attainment resulting from the training. The training outcome was measured using incremental knowledge related to training subjects. Socio-demographic-economic characteristics of training participants were considered the determinants of training outcomes. The analysis employed a multiple regression model. Data were compiled from the training centre’s reports and documentation of training held in 2021 and 2022. The results show that the training outcome was apparent. Some participants showed good knowledge acquisition, and some others did not. The starting point of existing knowledge mattered. Some socio-demographic-economic characteristics significantly determined incremental knowledge acquisition. The study suggests training participants should be grouped based on the knowledge baseline. The advanced participants are separated from the usual ones whenever possible to reach high training outcomes.

A Procedural Constructive Learning Mechanism with Deep Reinforcement Learning for Cognitive Agents

Recent advancements in AI and deep learning have created a growing demand for artificial agents capable of performing tasks within increasingly complex environments. To address the challenges associated with continuous learning constraints and knowledge capacity in this context, cognitive architectures inspired by human cognition have gained significance. This study contributes to existing research by introducing a cognitive-attentional system employing a constructive neural network-based learning approach for continuous acquisition of procedural knowledge. We replace an incremental tabular Reinforcement Learning algorithm with a constructive neural network deep reinforcement learning mechanism for continuous sensorimotor knowledge acquisition, thereby enhancing the overall learning capacity. The primary emphasis of this modification centers on optimizing memory utilization and reducing training time. Our study presents a learning strategy that amalgamates deep reinforcement learning with procedural learning, mirroring the incremental learning process observed in human sensorimotor development. This approach is embedded within the CONAIM cognitive-attentional architecture, leveraging the cognitive tools of CST. The proposed learning mechanism allows the model to dynamically create and modify elements in its procedural memory, facilitating the reuse of previously acquired functions and procedures. Additionally, it equips the model with the capability to combine learned elements to effectively adapt to complex scenarios. A constructive neural network was employed, initiating with an initial hidden layer comprising one neuron. However, it possesses the capacity to adapt its internal architecture in response to its performance in procedural and sensorimotor learning tasks, inserting new hidden layers or neurons. Experimentation conducted through simulations involving a humanoid robot demonstrates the successful resolution of tasks that were previously unsolved through incremental knowledge acquisition. Throughout the training phase, the constructive agent achieved a minimum of 40% greater rewards and executed 8% more actions when compared to other agents. In the subsequent testing phase, the constructive agent exhibited a 15% increase in the number of actions performed in contrast to its counterparts.

Knowledge Representation for Legal Document Summarization

This paper presents a novel approach for legal document summarization. Proposed approach is based on Ripple-Down Rules (RDR). It is an incremental knowledge acquisition method. RDR allows us to quickly build an extendable knowledge base using classification rules. The classification rules are written using a set of features. Summary is generated using the identified rhetorical roles in the document. Experiments demonstrate that the RDR based Legal Document summarization approach outperforms the supervised and unsupervised machine learning models.

Incremental Knowledge-based System for Recommending Content Adaptation in Dynamic Learning Environment

Dynamic learning environment (DLE) provides an opportunity for students with a remarkable learning experience in a limited time and specific situations. In this condition, adaptation and personalization have been key issues to accommodate differences between students. Both paradigms emphasize tailoring learning activities to students’ understanding and interest through learning objectives, instructional approaches, and learning pathways. In addition, the students will learn optimized instructional activities at their own pace. This paper presents an incremental knowledge-based system to facilitate learning content adaptation in DLE. To be specific, the knowledge base contains a set of rules incrementally constructed using Ripple Down Rules (RDR) after evaluating a series of test cases. The test cases are generated automatically by analyzing the attributes that reflect the learning situation. Since it is impossible to perform thorough testing involving all input parameters, the selection criteria using pairwise testing are applied to minimize the refinement. Therefore, the evaluation of the theoretical concept is then carried out on a real case. The selected case study for the analysis is the subject of Computer Networking for an undergraduate course. Several adaptive scenarios are presented based on some criteria. An education expert is involved in recommending suitable content for adaptation during the evaluation phase. However, the knowledge base development is automatically constructed from the incremental knowledge acquisition process. As the evaluation progresses, the knowledge base is validated for its accuracy in predicting learning content recommendations.

Parental Innovation Imprinting on CEO and Firm Knowledge Acquisition

In this paper we theorize that a firm’s knowledge acquisition will be shaped by its CEO’s experiences during his/her developmental years. Drawing on imprinting theory, we posit that CEOs whose parents had occupations requiring more innovation skills, will be more likely to engage in knowledge acquisitions compared to other CEOs. They acquire knowledge both externally - by acquiring other firms and internally - by investing more in research and development (R&D). Surprisingly, we find that these CEOs prefer incremental knowledge acquisition that is closely related to their firm’s existing knowledge. We also expect the main effects to be stronger following better firm performance and to weaken with CEO age. We find support for our hypotheses within a dataset of Fortune 100 firms between 2000 and 2016.

How does knowledge affect IPO pricing? A new approach grounded in the knowledge-based view

PurposeThis study aims to empirically validate that a knowledge-based view (KBV) is an important framework to understand price discovery processes in initial public offerings (IPOs) by emphasizing the unique feature of knowledge creation jointly invoked by underwriters and institutional investors during the book building phase.Design/methodology/approachThe authors decompose underwriters’ incremental knowledge acquisition into objective knowledge – acquired from premarket bids – and subjective knowledge – which is orthogonal to the objective knowledge. The authors implement a multiplicative heteroscedasticity model to analyze how each knowledge component relates to the level and volatility (as a proxy of pricing uncertainty) of post-issue returns. The authors take the 2007 regulatory change as a quasi-natural experiment in which institutional investors were incentivized to provide true information.FindingsFor Korean IPOs, the authors find that the objective (subjective) knowledge component reduces (increases) both pricing uncertainty and underpricing. The authors also observe that the efficacy of the IPO knowledge creation critically depends on the quality of the information provided by institutional investors, as anticipated by the KBV literature.Originality/valueUsing fine-grained knowledge measures, the authors provide original, compelling evidence that objective (subjective) knowledge formulated from the IPO knowledge-creation processes de facto alleviates (worsens) underwriters’ pricing difficulties. This reinforces the importance of knowledge-based mechanisms in managerial decision-making processes.

Knowledge Acquisition Approach Based on Incremental Objects From Data With Missing Values

Knowledge acquisition is the process of extracting useful knowledge from data sets to analyze data in areas of data mining and knowledge discovery. Most current knowledge acquisition work mainly focuses on static data. However, due to the dynamic characteristics of data, the objects grow at an unprecedented rate in real-world data sets. The incremental objects with a dynamic environment significantly affect knowledge updating. To maintain the effectiveness of knowledge from the dynamic data, it is necessary to update the knowledge timely. So far, there are relatively few studies on knowledge acquisition for the data with missing feature values, i.e., incomplete data. To handle with this issue, an incremental updating manner of the accuracy matrix and coverage matrix are first proposed on the basis of the computations of the tolerance classes in incomplete data, which plays an important role in the knowledge acquisition process. Then, an incremental knowledge acquisition algorithm is proposed when some new objects added to the data with missing values. Finally, some numerical experiments are conducted to evaluate the efficiency of the proposed algorithm.

Experts community memory for entity similarity functions recommendation

Similarity search (or similar entity search) is the process of finding all entities similar to a given entity (e.g., a person, a document, or an image). Although many techniques for similarity analysis have been proposed in the past, little work has been done on the question of which of the presented techniques are most suitable for a given similarity search task. Knowing the right similarity function is important as the task is highly domain- and data-dependent. In this article, we provide an approach for recommending which similarity functions (e.g., edit distance or jaccard similarity) should be used for measuring the similarity between two entities. The approach employs an incremental knowledge acquisition technique for capturing domain experts’ knowledge about similarity functions and their usage contexts (e.g., entity class, attribute name and some keywords). In addition, for situations where domain experts have little or no knowledge about datasets, we analyze the features of the datasets and then suggest similarity functions based on the identified features. We also demonstrate the feasibility and effectiveness of our proposed approach on several real-world datasets from different domains.

The Incremental Knowledge Acquisition Based on Hash Algorithm

In this paper, a new incremental knowledge acquisition method is proposed based on rough set theory, decision tree and granular computing. In order to effectively process dynamic data, describing the data by rough set theory, computing equivalence classes and calculating positive region with hash algorithm are analyzed respectively at first. Then, attribute reduction, value reduction and the extraction of rule set by hash algorithm are completed efficiently. Finally, for each new additional data, the incremental knowledge acquisition method is proposed and used to update the original rules. Both algorithm analysis and experiments show that for processing the dynamic information systems, compared with the traditional algorithms and the incremental knowledge acquisition algorithms based on granular computing, the time complexity of the proposed algorithm is lower due to the efficiency of hash algorithm and also this algorithm is more effective when it is used to deal with the huge data sets.

A robust transformation-based learning approach using ripple down rules for part-of-speech tagging

In this paper, we propose a new approach to construct a system of transformation rules for the Part-of-Speech (POS) tagging task. Our approach is based on an incremental knowledge acquisition method where rules are stored in an exception structure and new rules are only added to correct the errors of existing rules; thus allowing systematic control of the interaction between the rules. Experimental results on 13 languages show that our approach is fast in terms of training time and tagging speed. Furthermore, our approach obtains very competitive accuracy in comparison to state-of-the-art POS and morphological taggers.

CrashEd – A live immersive, learning experience embedding STEM subjects in a realistic, interactive crime scene

Interactive experiences are rapidly becoming popular via the surge of ‘escape rooms’; part game and part theatre, the ‘escape’ experience is exploding globally, having gone from zero offered at the outset of 2010 to at least 2800 different experiences available worldwide today. CrashEd is an interactive learning experience that parallels many of the attractions of an escape room – it incorporates a staged, realistic ‘crime scene’ and invites participants to work together to gather forensic evidence and question a witness in order to solve a crime, all whilst competing against a ticking clock. An animation can enhance reality and engage with cognitive processes to help learning; in CrashEd, it is the last piece of the jigsaw that consolidates the students’ incremental acquisition of knowledge to tie together the pieces of evidence, identify a suspect and ultimately solve the crime. This article presents the background to CrashEd and an overview of how a timely placed animation at the end of an educational experience can enhance learning. The lessons learned, from delivering bespoke versions of the experience to different demographic groups, are discussed. The article will consider the successes and challenges raised by the collaborative project, future developments and potential wider implications of the development of CrashEd. Keywords: animations; avatars; connectivism; higher education (Published: 20 September 2016) Citation: Research in Learning Technology 2016, 24 : 30089 - http://dx.doi.org/10.3402/rlt.v24.30089

Incremental Knowledge Acquisition for WSD: A Rough Set and IL based Method

Word sense disambiguation (WSD) is one of tricky tasks in natural language processing (NLP) as it needs to take into full account all the complexities of language. Because WSD involves in discovering semantic structures from unstructured text, automatic knowledge acquisition of word sense is profo

HAUSS: Incrementally building a summarizer combining multiple techniques

The idea of automatic summarization dates back to 1958, when Luhn invented the “auto abstract” (Luhn, 1958). Since then, many diverse automatic summarization approaches have been proposed, but no single technique has solved the increasingly urgent need for automatic summarization. Rather than proposing one more such technique, we suggest that the best solution is likely a system able to combine multiple summarization techniques, as required by the type of documents being summarized. Thus, this paper presents HAUSS: a framework to quickly build specialized summarizers, integrating several base techniques into a single approach. To recognize relevant text fragments, rules are created that combine frequency, centrality, citation and linguistic information in a context-dependent way. An incremental knowledge acquisition framework strongly supports the creation of these rules, using a training corpus to guide rule acquisition, and produce a powerful knowledge base specific to the domain. Using HAUSS, we created a knowledge base for catchphrase extraction in legal text. The system outperforms existing state-of-the-art general-purpose summarizers and machine learning approaches. Legal experts rated the extracted summaries similar to the original catchphrases given by the court. Our investigation of knowledge acquisition methods for summarization therefore demonstrates that it is possible to quickly create effective special-purpose summarizers, which combine multiple techniques, into a single context-aware approach.

Outcomes of longitudinal integrated clinical placements for students, clinicians and society

Longitudinal integrated clerkships (LICs) have been widely implemented in both rural and urban contexts, as is now evident in the wealth of studies published internationally. This narrative literature review aims to summarise current evidence regarding the outcomes of LICs for student, clinician and community stakeholders. Recent literature was examined for original research articles pertaining to outcomes of LICs. Students in LICs achieve academic results equivalent to and in some cases better than those of their counterparts who receive clinical education in block rotations. Students in LICs are reported to have well-developed patient-centred communication skills, demonstrate understanding of the psychosocial contributions to medicine, and report more preparedness in higher-order clinical and cognitive skills in comparison with students in traditional block rotations (TBRs). Students in LICs take on increased responsibility with patients and describe having more confidence in dealing with ethical dilemmas. Continuity of supervision reportedly facilitates incremental knowledge acquisition, and supervisors provide incrementally progressive feedback. Despite early disorientation regarding the organising of their learning, students feel well supported by the continuity of student-preceptor relationships and value the contributions made by these. Students in LICs living and working in rural areas are positively influenced towards primary care and rural career choices. A sound body of knowledge in the field of LIC research suggests it is time to move beyond descriptive or exploratory research that is designed to justify this new educational approach by comparing academic results. As the attributes of LIC alumni are better understood, it is important to conduct explanatory research to develop a more complete understanding of these findings and a foundation for new theoretical frameworks that underpin educational change. Longitudinal integrated clerkships are now recognised as representing credible and effective pedagogical alternatives to TBRs in medical education.

Interactive requirements prioritization using a genetic algorithm

ContextThe order in which requirements are implemented affects the delivery of value to the end-user, but it also depends on technical constraints and resource availability. The outcome of requirements prioritization is a total ordering of requirements that best accommodates the various kinds of constraints and priorities. During requirements prioritization, some decisions on the relative importance of requirements or the feasibility of a given implementation order must necessarily resort to a human (e.g., the requirements analyst), possessing the involved knowledge. ObjectiveIn this paper, we propose an Interactive Genetic Algorithm (IGA) that includes incremental knowledge acquisition and combines it with the existing constraints, such as dependencies and priorities. We also assess the performance of the proposed algorithm. MethodThe validation of IGA was conducted on a real case study, by comparing the proposed algorithm with the state of the art, interactive prioritization technique Incomplete Analytic Hierarchy Process (IAHP). ResultsThe proposed method outperforms IAHP in terms of effectiveness, efficiency and robustness to decision maker errors. ConclusionIGA produces a good approximation of the reference requirements ranking, requiring an acceptable manual effort and tolerating a reasonable human error rate.

Dynamic evaluation of the development process of knowledge-based information systems

Codifying expert domain knowledge is a difficult and expensive task. To evaluate the quality of the outcome, often the same domain expert or a colleague of similar expertise is relied on to undertake a direct evaluation of the knowledge-based system or indirectly by preparing appropriate test data. During an incremental knowledge acquisition process, a data stream is available, and the knowledge base is observed and amended by an expert each time it produces an error. Using the kept record of the system’s performance, we propose an evaluation process to estimate its effectiveness as it gets evolved. We instantiate this process for an incremental knowledge acquisition methodology, Ripple Down Rules. We estimate the added value in each knowledge base update. Using these values, the decision makers in the organisation employing the knowledge-based information system can apply a cost-benefit analysis of the continuation of the incremental knowledge acquisition process. They can then determine when this process, involving keeping an expert online, should be terminated. As a result, the expert is not kept on-line longer than it is absolutely necessary. Hence, a major expense in deploying the information system—the cost of keeping a domain expert on-line—is reduced.

COMBINING GRAPH-CUT TECHNIQUE AND ANATOMICAL KNOWLEDGE FOR AUTOMATIC SEGMENTATION OF LUNGS AFFECTED BY DIFFUSE PARENCHYMAL DISEASE IN HRCT IMAGES

Accurate and automated lung segmentation in high-resolution computed tomography (HRCT) is highly challenged by the presence of pathologies affecting lung parenchyma appearance and borders. The algorithm presented employs an anatomical model-driven approach and systematic incremental knowledge acquisition to produce coarse lung delineation, used as initialization for the graph-cut algorithm. The proposed method is evaluated on a 49 HRCT cases dataset including various lung disease patterns. The accuracy of the method is assessed using dice similarity coefficient (DSC) and shape differentiation metrics (d mean , d rms ), by comparing the outputs of automatic lung segmentations and manual ones. The proposed automatic method demonstrates high segmentation accuracy ( DSC = 96.64%, d mean = 1.75 mm, d rms = 3.27 mm) with low variation that depends on the lung disease pattern. It also presents good improvement over the initial lung segmentation (Δ DSC = 4.74%, Δd mean = -3.67 mm, Δd rms = -6.25 mm), including impressive amelioration (maximum values of Δ DSC = 58.22% and Δd mean = -78.66 mm) when the anatomy-driven algorithm reaches its limit. Segmentation evaluation shows that the method can accurately segment lungs even in the presence of disease patterns, with some limitations in the apices and bases of lungs. Therefore, the developed automatic segmentation method is a good candidate for the first stage of a computer-aided diagnosis system for diffuse lung diseases.

Ontology learning from biomedical natural language documents using UMLS

The generation of new knowledge is continuous in biomedical domains, thus biomedical literature is becoming harder to understand. Ontologies provide vocabulary standardization, so they can be helpful to facilitate the understanding of biomedical texts. In this work, a methodology for building biomedical ontologies from texts is presented. This approach relies on natural language processing and incremental knowledge acquisition techniques to obtain the relevant concepts and relations to be included in an OWL ontology. Additionally, we provide an algorithm to connect the isolated concepts regions in the ontology using UMLS. We also discuss in this paper the experiment carried out to validate our approach and its positive results in terms of performance and scalability.

Towards context-based learning for support systems

Often, decision-making processes are embedded in official procedures to address focuses in any case. However, procedures lead often to sub-optimal solutions for any specific decision making, and actors are obliged to develop practices to address the specificity of the context in which a decision is made. This opposition between procedure and practices is well known in different domains (prescribed and effective tasks, logic of functioning versus logic of use, etc.). We have shown what the differences were in a context-based formalism called Contextual Graphs. In this paper, we discuss the possibility to use of good and bad practices for the training of human actors, thanks to an incremental acquisition of knowledge and the learning of new practices by a system. We discuss these aspects in the framework of α real-world application in road safety (modeling of drivers' behaviors), but ideas can be easily reused in other domains.

MULTI-ENGINE COLLABORATIVE BOOTSTRAPPING FOR WORD SENSE DISAMBIGUATION

In this paper we propose a new word sense disambiguation method called Multi-engine Collaborative Bootstrapping (MCB) that combines different types of corpora and also uses two languages for bootstrapping. MCB uses the bilingual bootstrapping as its core algorithm that leading to incremental knowledge acquisition. The EM model is applied to train parameters in a base learner. The feature translation model is improved by semantic correlation estimation. In addition we use multi-engine selection to produce qualified starting seeds from parallel corpora and monolingual corpora. Those seeds that are generated through unsupervised machine learning approaches can also ensure bootstrapping effectiveness in contrast with manually selected seeds in spite of their different selection mechanisms. Experimental results prove the effectiveness of MCB. Some factors including feature space and starting seed number are concerned involved in our experiments because the EM algorithm is sensitive to starting values. Limitation of resources is also a concern.