Fuzzy systems and neural networks in software engineering project management

Kumar, Satish; Krishna, B. Ananda; Satsangi, P. S.

doi:10.1007/bf00872054

Cited by 70 publications

(32 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Fuzzy logic effort prediction brings numerous benefits. Undeniably the development of software is characterized by parameters that possess certain level of fuzziness [17]. This requires that some degree of uncertainty be introduced in the models, in order to make the models realistic.…”

Section: Why Fuzzy Logic?mentioning

confidence: 99%

Enhanced Software Development Effort and Cost Estimation Using Fuzzy Logic Model

Ling

Phang

et al. 2007

MJCS

View full text Add to dashboard Cite

Keywords: Fuzzy logic, cost estimation INTRODUCTIONSoftware metric refers to the measurement of software attributes which are typically related to the product, the process and the resources of software development [1]. These measurements can be used as parameters in project management models [2] which provide aids to software project managers in the daunting task of managing software projects to avoid problems such as cost overrun and behind the schedule.One of the most extensively researched areas of software measurement is software effort prediction. Software effort prediction models fall into two main categories: algorithmic and non-algorithmic. The most popular algorithmic prediction models include Boehm's COCOMO [3], Putnam's SLIM [4] and Albrecht's Function Point [5]. These models require as inputs, accurate estimate of certain attributes such as line of code (LOC), complexity and so on which are difficult to obtain during the early stage of a software development project. The models also have difficulty in modeling the inherent complex relationships between the contributing factors, are unable to handle categorical data as well as lack of reasoning capabilities [6]. The limitations of algorithmic models led to the exploration of the non-algorithmic techniques which are soft computing based. These include artificial neural network, evolutionary computation, fuzzy logic models, case-based reasoning and so on.Artificial neural network are used in effort estimation due to its ability to learn from previous data [7][8]. It is also able to model complex relationships between the dependent (effort) and independent variables (cost drivers) [7][8].In addition, it has the ability to generalize from the training data set thus enabling it to produce acceptable result for previously unseen data. Most of the work in the application of neural network to effort estimation made use of feed-forward multi-layer Perceptron, Backpropagation algorithm and sigmoid function [7].Despite of these, neural network's limitations in several aspects prevent it from being widely adopted in effort prediction [7]. It is a 'black box' approach and therefore it is difficult to understand what is going on internally within a neural network. Hence, justification of the prediction rationale is tough. Neural network is known of its ability in tackling classification problem. Contrarily, in effort estimation what is needed is generalization capability. At the same time, there is little guideline in the construction of neural network topologies. On the other hand, genetic programming was used in effort estimation by Burgess et al [9] and the use of case-based reasoning in effort estimation can be found in [8]. The work presented here focused on the application of fuzzy logic in effort prediction.

show abstract

Section: Why Fuzzy Logic?mentioning

confidence: 99%

Enhanced Software Development Effort and Cost Estimation Using Fuzzy Logic Model

Ling

Phang

et al. 2007

MJCS

View full text Add to dashboard Cite

show abstract

“…Using an experimental research method they found that a fuzzy logic approach was more effective at classifying mutated programs (as correct or incorrect) than a purely probabilistic approach. Kumar et al (1994) provide a well-reasoned justification for the use of fuzzy logic in software project management, building an illustrative fuzzy system to replicate Putnam's personnel scheduling model.…”

Section: Modeling Aspects Of Software Project Managementmentioning

confidence: 99%

“…After experimenting with various membership function shapes we restricted ourselves to trapezoidal and triangular as being appropriate for the discrete data being modeled (Kumar et al 1994). For the first of the two BUILD samples the best model (in terms of goodness-offit (see the next part of this section for further discussion)) was one that employed seven membership functions and fifteen rules.…”

Section: Fuzzy Modelingmentioning

confidence: 99%

“…The potential of such methods has been demonstrated in domains other than software engineering for some time, and in the last decade such methods have been increasingly adopted by software engineering researchers. For example, the capabilities of neural networks in analyzing software engineering data sets have been assessed among others by Khoshgoftaar and Lanning (1995), Lee et al (1998) and Mair et al (2000), case-based reasoning and analogy methods have been investigated by Shepperd and Schofield (1997), Myrtveit and Stensrud (1999) and Walkerden and Jeffery (1999), and fuzzy logic has been investigated by Kumar et al (1994), Ebert (1997), Yau and Tsoi (1998) and Piattini et al (2001). More recently, genetic algorithms have also been considered as potentially useful in analyzing software engineering data (Shukla 2000).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Software source code sizing using fuzzy logic modeling

MacDonell

2003

Information and Software Technology

View full text Add to dashboard Cite

show abstract

“…Even though a large number of different neural network architectures and training algorithms exist, almost all published studies involving software metric models have been limited to this type [26][27][28][29][30][31][32][33][34][35] . This can be seen as a reflection of the lack of understanding of neural network techniques by many software metric researchers which is understandable given the tremendous growth in the neural network field in the past decade.…”

Section: Neural Networkmentioning

confidence: 99%