ISSN (Online) 2278-1021 ISSN (Print) 2319-5940 International Journal of Advanced Research in Computer and Communication Engineering Vol. 4, Issue 2, February 2015 Copyright to IJARCCE DOI 10.17148/IJARCCE.2015.4286 380 A Comparison between Semantic and Syntactic Software Metrics Rasha Gaffer M. Helali Sudan University of Science and Technology, Khartoum, Sudan Abstract :Software metrics are quantitative estimates for software product attributes which guide us in taking managerial and technical decisions [1]. The current used software metrics depend on syntactical attributes of source code at the same time they ignore its semantic aspects. This fact motivates us to focus on semantic metrics instead of traditional used metrics. Moreover, semantic metric are more precise than syntactic ones, it is able to capture the semantic defects of the software products. In this paper, the researcher has been made an attempt to survey of existing research on software metrics along with potential research challenges and opportunities. Keywords: Metrics, Semantic, Synthetic, data mining. I. INTRODUCTION The quality of software systems become an important issue in continuation of business, i.e. The vast amount of software used in markets and their role in managing precise and dangerous tasks [1]. The assertion by Ward Cunningham in 1992, that quick and careless development with poor quality leads to many years of expensive maintenance and enhancements. So, software measurement paradigm becomes more and more important to monitor quality during all software development stages. The field of software metrics is relatively young one [2], whose origins can be found in the work by Halstead published in 1972. From then on, the interest in software metrics has increased because they have been recognized as a useful instrument for managing software process effectively. Software metrics allow to use a real engineering approach to software development, providing the quantitative and objective base that software engineering was lacking. In fact, their use in industry is becoming more and more widespread.[2] As regards the research in software metrics [2], it has undergone a great evolution, in the first period the focus was very much on inventing new metrics to measure different software attributes, without concerning the scientific validity of these metrics. Recently, much work has been done on how to apply measurement theory to software metrics and ensure their validity. [2] The remainder of this paper is organized as follows. In the next section characteristics and challenges in software measurement were discussed. Current used Metrics including syntactic and semantic metrics described in section 3. Section 4 examines the current Challenges and finally conclusion was presented II. CHARACTERISTICS AND CHALLENGES IN SOFTWARE MEASUREMENT A principal objective of software engineering is to improve the quality of software products. Many models spot distinction between tow types of software attributes internal and external ones [1]. Since internal attributes can be measured directly from source code, external attributes measured indirectly based on another internal attributes. Much research has been done on identifying quality models to support software quality improvements. Such quality models concentrated on mapping between internal and external attributes i.e. McCall, Boehm [4], FURPS(+) and ISO9126 (ISO 1991) models. The main objective of these models is to focus on formal products and identify key attributes of quality from the user prospective [5]. Boehm and MaCall model identified 3 key attributes called “quality factors” such as reliability, usability, maintainability (High level external attributes) but it related to many internal attributes called quality criteria [5] . Dromey and FURPS(+) [6] are focusing on the relationship between the quality attributes and the sub- attributes, as well as attempting to connect software product properties with software quality attributes. In 1992, derivation of McCall model was proposed as a basis for international standard software quality ISO9126 (ISO 1991). The model decomposes the quality into six factors as follow: functionality, efficiency, usability, maintainability and portability. Each of these factors is defined as a set of attributes. [5] Software Metrics are used to measure specific quality attributes using relationship identified by standard quality models. Current used metrics are succeed in measuring internal attributes but still there is shortcomings in measuring external attributes accurately such as reliability, availability …etc. The following section focus on current used metrics and their limitations. III. SOFTWARE METRICS In this section, we will discuss software metrics concept and classification along with their advantages and shortcomings each. According to the IEEE standard glossary of software engineering terms (adapted from [6]), metrics are a quantitative measure of the degree to which a system, component, or process possesses a given attribute."