A New Approach to Knowledge-Based Design of Recurrent Neural Networks

Kolman, Eyal; Margaliot, Michael

doi:10.1109/tnn.2008.2000393

Cited by 11 publications

(5 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This stage is critical in any handwriting recognition system. In other words, it is a transition from the observable space to the decision space of class membership [4].…”

Section: Hindi Character Recognitionmentioning

confidence: 99%

“…Mathematical morphology uses the structural element [4] to process images in the form of specific forms already chosen, typically less than the image, that function as a result operator on an image. The shape of the structural element, size, and orientation are selected according to previous knowledge by the relevant geometrical structures existing in the picture and the goal of the morphological operation conducted [4].…”

Section: Morphological Operationsmentioning

confidence: 99%

“…In SIF, the phrase is vectorized using FastText, a word model learned in common crawl and Wikipedia. Kolman et al [4] proposed an alternative method, termed a SIF described in Algorithm 1.…”

Section: Sentence Encoder After Pre-processingmentioning

confidence: 99%

See 2 more Smart Citations

Support Vector Machine Based Handwritten Hindi Character Recognition and Summarization

Dhankhar¹,

Gupta²,

Memon³

et al. 2022

Computer Systems Science and Engineering

View full text Add to dashboard Cite

In today's digital era, the text may be in form of images. This research aims to deal with the problem by recognizing such text and utilizing the support vector machine (SVM). A lot of work has been done on the English language for handwritten character recognition but very less work on the under-resourced Hindi language. A method is developed for identifying Hindi language characters that use morphology, edge detection, histograms of oriented gradients (HOG), and SVM classes for summary creation. SVM rank employs the summary to extract essential phrases based on paragraph position, phrase position, numerical data, inverted comma, sentence length, and keywords features. The primary goal of the SVM optimization function is to reduce the number of features by eliminating unnecessary and redundant features. The second goal is to maintain or improve the classification system's performance. The experiment included news articles from various genres, such as Bollywood, politics, and sports. The proposed method's accuracy for Hindi character recognition is 96.97%, which is good compared with baseline approaches, and system-generated summaries are compared to human summaries. The evaluated results show a precision of 72% at a compression ratio of 50% and a precision of 60% at a compression ratio of 25%, in comparison to state-of-the-art methods, this is a decent result.

show abstract

“…This stage is critical in any handwriting recognition system. In other words, it is a transition from the observable space to the decision space of class membership [4].…”

Section: Hindi Character Recognitionmentioning

confidence: 99%

Section: Morphological Operationsmentioning

confidence: 99%

See 1 more Smart Citation

Support Vector Machine Based Handwritten Hindi Character Recognition and Summarization

Dhankhar¹,

Gupta²,

Memon³

et al. 2022

Computer Systems Science and Engineering

View full text Add to dashboard Cite

show abstract

“…A. Dickerson used a fast simulated annealing to reduce the number of fuzzy rules [5]. E. Kolman and M. Margaliot proposed a fuzzy rule base with a special structure, referred to the Fuzzy All-permutations Rule Base (FARB) [6]. M. Mizumoto and Y. Shi presented a learning algorithm for tuning fuzzy rules [7].…”

Section: Introductionmentioning

confidence: 99%

Development Multiple Neuro-Fuzzy System Using Back-propagation Algorithm

Alshawi

Allamy

Khan

2013

IJMIT

View full text Add to dashboard Cite

When fuzzy systems are highly nonlinear or include a large number of input variables, the number of fuzzy rules constituting the underlying model is usually large. Dealing with a large-size fuzzy model may face many practical problems in terms of training time, ease of updating, generalizing ability and interpretability. Multiple Fuzzy System (MFS) is one of effective methods to reduce the number of rules, increase the speed to obtain good results. This paper is therefore proposes another approach call Multiple Neuro-Fuzzy System (MNFS) which can further enhance the performance of the MFS approach. The new approach is used Back-propagation algorithm in the learning process. The performance of the proposed approach evaluates and compares with MFS by three experiments on nonlinear functions. Simulation results demonstrate the effectiveness of the new approach than MFS with regards to enhancement of the accuracy of the results.

show abstract

“…We focus on extracting symbolic knowledge, stated as a suitable FRB, from trained RNNs, leaving the issue of knowledge insertion to a companion paper [28]. We demonstrate the usefulness of our approach by applying it to provide a comprehensible description of the functioning of an RNN trained to recognize a formal language.…”

Section: Introductionmentioning

confidence: 99%

Extracting symbolic knowledge from recurrent neural networks—A fuzzy logic approach

Kolman

Margaliot

2009

Fuzzy Sets and Systems

View full text Add to dashboard Cite

Considerable research has been devoted to the integration of fuzzy logic (FL) tools with classic artificial intelligence (AI) paradigms. One reason for this is that FL provides powerful mechanisms for handling and processing symbolic information stated using natural language. In this respect, fuzzy rule-based systems are white-boxes, as they process information in a form that is easy to understand, verify and, if necessary, refine.The synergy between artificial neural networks (ANNs), which are notorious for their black-box character, and FL proved to be particularly successful. Such a synergy allows combining the powerful learning-from-examples capability of ANNs with the high-level symbolic information processing of FL systems.In this paper, we present a new approach for extracting symbolic information from recurrent neural networks (RNNs). The approach is based on the mathematical equivalence between a specific fuzzy rulebase and functions composed of sums of sigmoids. We show that this equivalence can be used to provide a comprehensible explanation of the RNN functioning. We demonstrate the applicability of our approach by using it to extract the knowledge embedded within an RNN trained to recognize a formal language.

show abstract

A New Approach to Knowledge-Based Design of Recurrent Neural Networks

Cited by 11 publications

References 62 publications

Support Vector Machine Based Handwritten Hindi Character Recognition and Summarization

Support Vector Machine Based Handwritten Hindi Character Recognition and Summarization

Development Multiple Neuro-Fuzzy System Using Back-propagation Algorithm

Extracting symbolic knowledge from recurrent neural networks—A fuzzy logic approach

Contact Info

Product

Resources

About