Article Preview
Top1. Introduction
The need of software in numerous applications has substantially risen in the last two decades (Feng et al., 2021a). To satisfy the customer's requirements, a large number of software applications for daily usage or business are established. Because of the mass manufacturing of software packages, the software quality has remained an unsolved issue resulting in inadequate performance for both personal as well as commercial applications (Eken & Tosun, 2021). To resolve this concern, software testing was developed that aids in the detection of bugs or faults in software applications and attempts to repair them. Due to the rising demand from commercial applications and modern technology-based homes, a sizable variety of new applications are produced each year; nonetheless, software quality has remained a neglected issue throughout this development (Zhao et al., 2019). In recent years, software applications have expanded significantly for both everyday use and business purposes. A little software error in a commercial context might harm the sector and lower consumer happiness. Software testing challenges (Majd et al., 2020; Sun et al., 2020) have grown to be a significant source of worry. As a result, a software analysis technique that calls for an automatic procedure may be more effective and need less installation money. If the tester is aware of the causes of probable flaws and the whole software development process, this issue may be avoided (Qiao et al., 2020). This might help with enhancing the strategy and execution of software development projects in addition to decreasing the overall expenses of software development (Ni et al., 2019).
In the software industry, the development methodology is recognised as a key contribution of programme design. Project managers are more frequently asking for early defect forecasts (Xu, Liu, & Zhang, 2019), and they see it as a crucial but difficult responsibility. This is in line with current trends. Recent advancements in the software development industry include challenging problem domains, an increasing complex design process, unpredictable software performance, and software application requirements. Although there is thorough documenting and an organized technique, certain errors are unavoidable throughout development of software process (Xu, Li, & Tang, 2019) (Gollagi et al., 2019), which negatively impacts software performance (Cai et al., 2021). In the current industrial expansion, several techniques for software fault minimization are introduced. However, these approaches require more time, money, and resources in order to analyse software applications properly. The study of failure causes and the improvement of software performance may benefit from these efforts (Gollagi, Math, & Daptardar, 2020). Additionally, in accordance with software dependability (Zhou et al., 2018), the software appliances are assessed in a certain inconsistent environment and tested to determine whether they can function there for a predetermined amount of time. This approach is focused on estimating the likelihood of software dependability. This work makes use of software measurements, and it is discovered that a significant number of failure forecasts require extra performance improvement resources, making the procedure challenging. Several SDP (Shao et al., 2020) (Gollagi, Pareek, & Pai H, 2020) techniques that use software metrics have been put out in recent years. These models could help in the development of reliable software and the early identification of flaws. The majority of defect prediction models that have been developed to far have relied primarily on software data. Software measures are utilized to build a statistics model for predicting software issues in a different situation for SDP (Zhu et al., 2021) (Gollagi, Piyush, Pareek et al, 2020).