Case-based reasoning research and development : 29th International Conference, ICCBR 2021, Salamanca, Spain, September 13-16, 2021 : proceedings / / Antonio A. Sánchez-Ruiz, Michael W. Floyd

• Autore: Sánchez-Ruiz Antonio A.
• Pubblicazione: Cham, Switzerland : , : Springer International Publishing, , [2021]; ©2021
• Descrizione fisica: 1 online resource (337 pages)
• Disciplina: 006.33
• Soggetto topico: Expert systems (Computer science); Case-based reasoning
• Persona (resp. second.): FloydMichael W.
• Nota di contenuto: Intro -- Preface -- Organization -- Contents -- The Bites Eclectic: Critique-Based Conversational Recommendation for Diversity-Focused Meal Planning -- 1 Introduction -- 2 Background -- 2.1 Diversity -- 2.2 Critiquing -- 2.3 Recipe Recommendation and Diversity -- 3 DiversityBite: A Conversational Dynamic Critique-Based Recommender -- 3.1 Recipe Case Representation and Similarity -- 3.2 Diversity-Focused Conversational Critique -- 4 Evaluation for DiversityBite -- 4.1 Recipe Dataset -- 4.2 Evaluation Study -- 4.3 Evaluation Results -- 5 Discussion and Future Work -- References -- Evaluation of Similarity Measures for Flight Simulator Training Scenarios -- 1 Introduction -- 2 Related Work -- 3 Problem Definition: Pilot Competencies and Event Sets -- 4 Case Retrieval of Scenario Event Sets -- 5 Evaluation of Similarity Measures -- 5.1 Experimental Setup -- 5.2 Analysis -- 6 Results and Discussion -- 7 Conclusions -- References -- Instance-Based Counterfactual Explanations for Time Series Classification -- 1 Introduction -- 2 Related Work -- 3 Good Counterfactuals for Time Series: Key Properties -- 4 Native Guide: Counterfactual XAI for Time Series -- 5 Testing Native Guide: Two Comparative Experiments -- 5.1 Experiment 1: Probing Proximity and Sparsity -- 5.2 Experiment 2: Exploring Plausibility and Diversity -- 6 Conclusion and Future Directions -- References -- User Evaluation to Measure the Perception of Similarity Measures in Artworks -- 1 Introduction -- 2 Related Work About Similarity -- 3 Methodology for Learning Similarity Measures Reflecting Human Perceptions -- 3.1 Methodology for Learning Perception Aware Similarity Measures -- 3.2 Profile Definition -- 4 Experiment on the Perception of Similarity for Artworks -- 4.1 Definition of Local Similarity Measures -- 4.2 Data Gathering of Perceived Similarity -- 4.3 Experimental Results. ; 5 Conclusions and Future Work -- References -- Measuring Financial Time Series Similarity with a View to Identifying Profitable Stock Market Opportunities -- 1 Introduction -- 2 Related Work -- 3 From Prices to Cases -- 4 Similarity in Financial Time Series -- 4.1 The Problem with Correlation -- 4.2 An Adjusted Correlation Metric -- 4.3 A Novel Similarity Metric for Returns-Based Time-Series -- 4.4 Most and Least Similar Cases -- 5 Evaluation -- 5.1 Predicting Monthly Returns -- 5.2 Comparing Trading Strategies -- 6 Conclusion and Future Work -- References -- A Case-Based Reasoning Approach to Predicting and Explaining Running Related Injuries -- 1 Introduction -- 2 Related Work -- 3 CBR for Injury Prediction and Explanation -- 3.1 Representing Training Load -- 3.2 Representing Injury Cases -- 3.3 Balancing the Case Base -- 3.4 Task 1: Predicting Training Breaks -- 3.5 Task 2: Explaining Training Breaks -- 4 Evaluation -- 4.1 Setup -- 4.2 Evaluating Prediction Accuracy -- 4.3 Evaluating Injury Explanations -- 5 Conclusions -- References -- Bayesian Feature Construction for Case-Based Reasoning: Generating Good Checklists -- 1 Introduction -- 2 Related Work -- 3 Case and Problem Definition -- 4 BCBR Framework -- 4.1 Bayesian Inference -- 4.2 Case Base Creation and CBR Engine -- 4.3 Example: NBI Estimates, Case Retrieval and CBR Case -- 5 Experiments -- 5.1 Experimental Setup -- 5.2 Experiment 1: Answer Classification Performance (Baselines) -- 5.3 Experiment 2: Trustworthiness of Constructed Checklists -- 5.4 Experiment 3: Evaluation of Constructed Checklists -- 6 Conclusion -- References -- Revisiting Fast and Slow Thinking in Case-Based Reasoning -- 1 Introduction -- 2 Background -- 3 Feature Selection -- 3.1 Feature Selection for Parsimonious Search -- 3.2 Modelling Desirability of Feature Sets -- 3.3 Brute Force Approach -- 3.4 Greedy Approach. ; 3.5 Experimental Analysis -- 4 Constraint-Based Switching -- 4.1 Model 2D -- 4.2 Experimental Analysis -- 5 Complexity Measure for Dichotomous Models -- 5.1 Footprint Size -- 5.2 Footprint with Time -- 5.3 Experimental Analysis -- 6 Conclusion and Future Work -- References -- Harmonizing Case Retrieval and Adaptation with Alternating Optimization -- 1 Introduction -- 2 Background -- 3 Alternating Optimization of Retrieval and Adaptation -- 4 Testbed System Design -- 4.1 Loss Function -- 4.2 Testbed Retrieval and Adaptation -- 4.3 Testbed Training and Testing Procedures -- 5 Evaluation -- 5.1 Experimental Settings -- 5.2 Experimental Results -- 6 Guidelines for Applying AO to Train CBR Components -- 7 Future Work -- 8 Conclusion -- References -- Adaptation Knowledge Discovery Using Positive and Negative Cases -- 1 Introduction -- 2 Motivations and Preliminaries -- 2.1 Assumptions and Notations About CBR -- 2.2 Boolean Setting Illustrated with a Boolean Function Example -- 2.3 Itemset Extraction -- 3 Exploiting Case Variations for Adaptation Knowledge Discovery with Positive and Negative Cases -- 4 Experiments on Benchmarks -- 4.1 Experiment Setting and Evaluation Methodology -- 4.2 Congressional Voting Records -- 4.3 Tic Tac Toe Endgame -- 4.4 Cardiac Diagnosis -- 4.5 Car Evaluation -- 4.6 Results and Discussion -- 5 Conclusion -- References -- When Revision-Based Case Adaptation Meets Analogical Extrapolation -- 1 Introduction -- 2 Setting of the Problem and Running Example -- 2.1 A Quick Refresher About Propositional Logic -- 2.2 Notions and Notations Related to CBR -- 2.3 Specification of the Running Example -- 2.4 Analogical Proportions and CBR -- 2.5 Belief Revision and CBR -- 3 Bridging Extrapolation and Revision-Based Adaptation -- 3.1 Reformulating Adaptation by Extrapolation as a Single Case Adaptation. ; 3.2 A Revision Operator Based on Competence of Case Pairs -- 3.3 An Approach to Adaptation Based on Extrapolation and Revision -- 3.4 Synthesis -- 4 Related Work and Final Remarks -- References -- Inferring Case-Based Reasoners' Knowledge to Enhance Interactivity -- 1 Introduction -- 2 Problem Statement: Interaction with a CBR Agent -- 3 Modeling the User as a Case-Based Reasoner -- 4 Inference of the CBR Parameters -- 4.1 General Principle -- 4.2 Inference of the Parameters for a Deterministic CBR -- 4.3 Probability of Retrieval for kNN -- 4.4 Discussion on the Inference Process -- 5 Application: Teaching Word Inflection -- 5.1 Presentation of the Application -- 5.2 Implementation of a Case-Based Reasoning Learner -- 5.3 Empirical Evaluation -- 6 Conclusion -- References -- A Case-Based Approach for the Selection of Explanation Algorithms in Image Classification -- 1 Introduction -- 2 Background -- 2.1 LIME -- 2.2 Anchors -- 2.3 Integrated Gradients -- 2.4 XRAI -- 3 CBR Process Specification -- 3.1 Case Base Elicitation -- 3.2 Similarity Metrics -- 3.3 Reuse Strategies -- 4 Evaluation -- 5 Conclusions and Future Work -- References -- Towards Richer Realizations of Holographic CBR -- 1 Introduction -- 2 Holographic Case-Based Reasoning -- 3 Methodology -- 3.1 Key Ideas -- 3.2 Holographic CBR Realization Framework -- 4 Results and Interpretations -- 4.1 Comparison with Baseline -- 4.2 Tests for Efficiency -- 5 Conclusions and Future Directions -- References -- Handling Climate Change Using Counterfactuals: Using Counterfactuals in Data Augmentation to Predict Crop Growth in an Uncertain Climate Future -- 1 Introduction -- 1.1 The Problem: Grass Growth Prediction for Sustainable Dairy Farming -- 1.2 Related Work: Counterfactuals from XAI to Data Augmentation -- 1.3 Research Questions and Novelties. ; 2 Study 1: Predicting Climate Disruption with PBI-CBR -- 2.1 Defining a Class Boundary for Climate Outlier Cases -- 2.2 Experiment 1a: The Contribution of Climate Outliers to Predictions -- 2.3 Experiment 1b: Role of Training Outliers at Values of k -- 3 Study 2: Predicting Climate Disruption with Counterfactuals -- 3.1 A Case-Based Counterfactual Augmentation Algorithm (CFA) -- 3.2 Experiment 2: Using Synthetic Counterfactual Cases to Predict Growth -- 4 Conclusions: Novelties, Explications and Caveats -- References -- A Case-Based Approach to Data-to-Text Generation -- 1 Introduction -- 2 Related Works -- 3 Background -- 4 Methodology -- 4.1 Case-Base Creation -- 4.2 Retrieval and Feature Weighting -- 4.3 Generation -- 5 Experimental Setup -- 5.1 Dataset -- 5.2 Baseline and Benchmark -- 5.3 Evaluation Methods -- 6 Results and Discussion -- 6.1 Comparison with Benchmark and Baseline -- 6.2 Ablation Studies -- 6.3 Qualitative Analysis -- 7 Conclusion and Future Work -- References -- On Combining Knowledge-Engineered and Network-Extracted Features for Retrieval -- 1 Introduction -- 2 Convolutional Neural Networks for Classification -- 3 Related Work -- 4 Bridging Engineered and Network-Extracted Features -- 5 Evaluation -- 5.1 Test Domain and Testbed System -- 5.2 Preliminary Experiments to Set Network Parameters -- 5.3 How Retrieval Accuracy Changes with KE Feature Degradation -- 5.4 How Using KE and NL Features in Concert Affects Accuracy -- 5.5 How Learned Weights Further Influence Retrieval Accuracy -- 6 Ramifications for Explainability -- 7 Conclusions -- References -- Task and Situation Structures for Case-Based Planning -- 1 Introduction -- 2 Task Structure and Planning -- 2.1 Task Structure -- 2.2 Execution of Tasks -- 2.3 Implementation of Our Task Structure -- 2.4 Discussions on Task Structure -- 3 Situation Structure -- 4 Situation Handling. ; 5 Illustrative Examples.
• Titolo autorizzato: Case-based reasoning research and development
• ISBN: 3-030-86957-1
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910502645803321
• Lo trovi qui: Univ. Federico II
• Serie: Lecture Notes in Computer Science