China
Africa
Explore chapters and articles related to this topic
AI-Informed Analytics Cycle: Reinforcing Concepts
Published in Jay Liebowitz, Data Analytics and AI, 2020
Rosina O. Weber, Maureen P. Kinkela
The first wave AI methods can roughly be categorized as data-oriented, knowledge-oriented, natural language, and evolutionary, as shown in Figure 12.4. Data-oriented methods typically learn from data and characterize the subfield of AI called machine learning. Machine learning methods (Russell and Norvig, 2014) include neural networks, decision trees, naive Bayes and support vector machines, and so on. These methods require large data sets to work well. Knowledge-based methods include rule-based reasoning, constraint satisfaction, and ontologies. Case-based reasoning combines both knowledge-based and machine learning aspects. Knowledge-based methods require some type of knowledge engineering and maintenance. Natural language processing combines techniques to treat sentences and grammar (ibid.), whereas natural language understanding theorizes the existence of a language model that can be used to process natural language segments (Allen, 1995). An example of an evolutionary method is genetic algorithms (Russell and Norvig, 2014), which evolve solution representations. This type of method is usually combined with others for best results. It requires a reference of quality as a fitness function to guide the evolution.
Establish knowledge processing (framework)
Published in Lukasz Rosinski, Knowledge Management for Project Excellence, 2019
‘Case-based Reasoning’ (CBR) is a mechanism to support direction, and thereby knowledge application. According to Watson (2001), CBR is not an AI technology such as logic programming, rule-based reasoning or neural computing, but instead it is a methodology for problem solving. More specifically, Case-based reasoning is a problem-solving paradigm that is able to utilize the specific knowledge of previously experienced, concrete situations (Ribeiro, 2005, p. 3). A key feature of CBR is its coupling to learning. Ribeiro (2005) explains that learning from experience in CBR happens as a natural by-product of problem solving. He also adds that learning is an intrinsic part of CBR because the solutions to past problems and their outcomes are stored as cases to extend the reasoner’s knowledge.
Telecommunications
Published in Jay Liebowitz, The Handbook of Applied Expert Systems, 2019
Both machine learning techniques take a set of training examples/cases as input. Each training example/case consists of a set of decision choices and a corresponding class. The decision tree approach constructs a decision tree that has internal nodes labeled as decisions and the leaves labeled as classes. At run time, a classification is done by following a path from the root node to a leaf, given only a set of decision choices. In contrast, the case-based reasoning approach just stores training cases in a memory. It classifies a new case by comparing it with stored cases at run time using a variety of indexing and matching strategies. The class of the most similar case is determined as the classification. These two machine learning techniques can be used to acquire domain knowledge through collecting training examples, and complement the direct approach of rule acquisition. Unlike neural networks, their reasoning processes can be analyzed symbolically. Note that each technique has its own specific classification bias (e.g., the decision tree approach is sensitive to the order of decision choices).
Emergency Management Case-Based Reasoning Systems: A Survey of Recent Developments
Published in Journal of Experimental & Theoretical Artificial Intelligence, 2023
Walid Bannour, Ahmed Maalel, Henda Hajjami Ben Ghezala
Derived from artificial intelligence, case-based reasoning is a problem-solving paradigm that mimics the way humans think to solve problems (Aamodt & Plaza, 1994). According to (Breslow & Aha, 1998), humans usually make decisions by comparing their current problematic situation to similar ones from the past. In CBR, a new problem is solved by remembering a similar previous case and reusing it in the new problematic situation (Aamodt & Plaza, 1994). A case means a prior problematic situation (i.e., an experience of a solved problem (Richter & Weber, 2016)) and is represented by a set of attributes reflecting the problem instance and its solution (F. Yu et al., 2018). A case can be represented in different forms: structured representation using frames or ontologies for instance, feature-value vector representation or textual representation (Bergmann et al., 2005). A set of organised and stored cases constitute a case base.
A maritime safety on-board decision support system to enhance emergency evacuation on ferryboats
Published in Maritime Policy & Management, 2019
Peiman Alipour Sarvari, Emre Cevikcan, Metin Celik, Alp Ustundag, Bilal Ervural
As the fifth phase of the methodology, a three-module DSS for MEE is developed. Since the modules of the DSS are developed using different methods, it will be meaningful to give introductory information about these techniques. As the first method used within the DSS, a heuristic algorithm is a set of general rules that are independent of the particular data case being considered in solving the problem (Martí and Reinelt 2011). Case-Based Reasoning, the next method considered in the DSS, is similar to the decision-making mechanism used by humans (Watson and Marir 1994). Case-Based Reasoning involves two main steps, namely finding similar cases in memory and adapting previous solutions to current problems (Aamodt and Plaza 1994; Teng, Chen, and Xia 2016). The other method used in the DSS (i.e. Rule-Based Reasoning) generates a set of outputs by applying a set of IF-THEN rules based on the information from a scenario (Yudin and Karpov 2017).
An integrated approach for automated physical architecture generation and multi-criteria evaluation for complex product design
Published in Journal of Engineering Design, 2019
Ruirui Chen, Yusheng Liu, Hongri Fan, Jianjun Zhao, Xiaoping Ye
Case-based reasoning refers to solving problems, evaluating solutions, explaining abnormal situations and understanding new situations by using old cases or experiences (Peng 2007). In this study, case-based mapping is inspired by case-based reasoning. A case is represented as shown in Figure 4. Here, a function corresponds to the ‘problem’ item, and a component corresponds to the ‘solution’ item. There exists additional information (the ‘situation’ item) in the case, which consists of the pros and cons of the solution (component) in solving (realising) the problem (function) in the previous products. When a component that is found in case-based mapping is expected to be used in the product, the information that is contained in the ‘situation’ item of its case can be used as a reference to decide whether the component should be used.