Inductive neural logic network and the SCM algorithm

Dounias²,

Panagi³

et al. 2006

Oncol Rep

Abstract. The present work proposes a computer assisted methodology for the effective modelling of the diagnostic decision for breast tumor malignancy. The suggested approach is based on innovative hybrid computational intelligence algorithms properly applied in related cytological data contained in past medical records. The experimental data used in this study, were gathered in the early 1990s in the University of Wisconsin, based in post diagnostic cytological observations performed by expert medical staff. Data were properly encoded in a computer database and accordingly, various alternative modelling techniques were applied on them, in an attempt to form diagnostic models. Previous methods included standard optimisation techniques, as well as artificial intelligence approaches, in a way that a variety of related publications exists in modern literature on the subject. In this report, a hybrid computational intelligence approach is suggested, which effectively combines modern mathematical logic principles, neural computation and genetic programming in an effective manner. The approach proves promising either in terms of diagnostic accuracy and generalization capabilities, or in terms of comprehensibility and practical importance for the related medical staff.

Section: Methodsmentioning

confidence: 99%

Breast tumor malignancy modelling using evolutionary neural logic networks

Dounias²,

Panagi³

et al. 2006

Oncol Rep

“…In [28], a training methodology related to back-propagation was proposed. Later, the Supervised Clustering and Matching (SCM) algorithm [27] was introduced. These training models however, aiming at the refinement of the edge weights often made the neural logic networks suffer in terms of their interpretability.…”

Section: Introductionmentioning

confidence: 99%

An evolutionary system for neural logic networks using genetic programming and indirect encoding

Journal of Applied Logic

Aggelis

Karkazis

et al. 2004

Nowadays, intelligent connectionist systems such as artificial neural networks have been proved very powerful in a wide area of applications. Consequently, the ability to interpret their structure was always a desirable feature for experts. In this field, the neural logic networks (NLN) by their definition are able to represent complex human logic and provide knowledge discovery. However, under contemporary methodologies, the training of these networks may often result in non-comprehensible or poorly designed structures. In this work, we propose an evolutionary system that uses current advances in genetic programming that overcome these drawbacks and produces neural logic networks that can be arbitrarily connected and are easily interpretable into expert rules. To accomplish this task, we guide the genetic programming process using a context-free grammar and we encode indirectly the neural logic networks into the genetic programming individuals. We test the proposed system in two problems of medical diagnosis. Our results are examined both in terms of the solution interpretability that can lead in knowledge discovery, and in terms of the achieved accuracy. We draw conclusions about the effectiveness of the system and we propose further research directions.

“…Even though powerful in their definition, neural logic networks are not widely applied. The main reason can be located in the fact that for the known training methodologies [1]- [2], the refinement of the edge weights reduces significantly the interpretability of these networks to expert rules, thus depriving these networks from their valuable feature.…”

Section: A Neural Logic Networkmentioning

confidence: 99%

“…Moreover, there is about 5% of missing values, and a combination of binary and continuous input features. Since the data is composed by features having different value ranges, we preprocessed the data by normalizing them into the [1,2] interval in order to be able to apply the fuzzy neural logic networks model. Specifically, using this scheme, a value of 1 denotes the 'false', a value of 2 denotes the 'true', while a value of 0 signifies the 'unknown'.…”

Section: B Diagnosis Of the Course Of Hepatitis Patientsmentioning

confidence: 99%

See 1 more Smart Citation

Towards neural-symbolic integration: the evolutionary neural logic networks

2004 2nd International IEEE Conference on 'Intelligent Systems'. Proceedings (IEEE Cat. No.04EX791)

AbstractThis work presents the application of a new methodology for the production of neural logic networks into two real-world problems from the medical domain. Namely, we apply grammar guided genetic programming using cellular encoding for the representation of neural logic networks into population individuals. The application area is consisted of the diagnosis of Diabetes and the diagnosis of the course of Hepatitis patients. The system is proved able to generate arbitrarily connected and interpretable evolved solutions leading to potential knowledge extraction.