Evaluating Explanations
نویسنده
چکیده
Explanation-based learning (EBL) is a powerful method for category formation. However, EBL systems are only effective if they start with good explanations. The problem of evaluating candidate explanations has received little attention: Current research usually assumes that a single explanation will be available for any situation, and that this explanation will be appropriate. In the real world many explanations can be generated for a given anomaly, only some of which are reasonable. Thus it is crucial to be able to distinguish between good and bad explanations. In people, the criteria for evaluating explanations are dynamic: they reflect context, the explainer’s current knowledge, and his needs for specific information. I present a theory of how these factors affect evaluation of explanations, and describe its implementation in ACCEPTER, a program to evaluate explanations for anomalies detected during story understanding.
منابع مشابه
A Critical Study on Three Explanations about the Real State of “Ehical Ought to”
Avoiding the ethical relativism and providing the preparations of ethical education are among the most important concerns of scholars in the field of ethics. Two theories of i`tibariat and realism about ethical cases of ought to are discussed among Muslim scholars each seeking to overcome the problem of ethical relativism. It seems that attempts for illustrating the ethical cases of ought to an...
متن کاملThe KNIGHT Experiments: Empirically Evaluating an Explanation Generation System
Empirically evaluating explanation generators poses a notoriously difficult problem. To address this problem, we constructed KNIGHT, a robust explanation generator that dynamically constructs natural language explanations about scientific phenomena. We then undertook the most extensive and rigorous empirical evaluation ever conducted on aa explanation generator. First, KNIGHT constructed explan...
متن کاملContext effects in explanation evaluation
Explanation is a critical component of higher cognitive functioning. We investigated whether some explanations are inherently appealing, or whether their appeal depends on context. To do so we presented statements (e.g., “Ducks have property X”) coupled with formal, causal, or teleological explanations (“because property X helps them float”), and asked participants to evaluate the explanations....
متن کاملThe Concept of Quality in Public Courtyards: Explanations and Analyses. Case Study: Mausoleum of Shah Ni’mat-Allah Vali
Quality is a highly esoteric concept which compels theorists to offer different explanations. Based on library resources, quality can be defined as an interaction between individuals and their environment, which is caused by a set of environmental components differing in each environment. This paper studied the concept of quality in public courtyards. The Mausoleum of Shah Ni’mat-Allah Vali, wh...
متن کاملEvaluating neural network explanation methods using hybrid documents and morphological prediction
We propose two novel paradigms for evaluating neural network explanations in NLP. The first paradigm works on hybrid documents, the second exploits morphosyntactic agreements. Neither paradigm requires manual annotations; instead, a relevance ground truth is generated automatically. In our experiments, successful explanations for Long Short Term Memory networks (LSTMs) were produced by a decomp...
متن کامل