Using Suffix Tree Clustering Method To Support The Planning Phase Of Systematic Literature Review

Abstract

Systematic Literature Review (SLR) is an information-based process throughout which each stage must be carefully and systematically designed in planning phase. Important decisions need to be made about the various choices involved in the SLR planning phase. However, due to the necessarily comprehensive and rigorous nature of SLR, many researchers have difficulties in planning SLRs sufficiently and effectively. In recent decades, the use of text mining (TM) techniques has been introduced to facilitate SLR process, especially at conducting review phase to support the searching and selection of studies. However, so far, TM techniques have not been applied to support any activities during SLR planning phase. In this research, we proposed a method that apply Suffix Tree Clustering (STC) method to support decision-making activities within the SLR planning phase. This method also known as SLR Planning Based on Suffix Tree Clustering (SLRP-STC) method. It aims to help reviewers in three ways: 1) to extract research topics easily; 2) to better interpret extracted information; and 3) to quickly realize and refine a proposed search string that is poorly formulated or inappropriate. A case study was conducted by comparing the proposed method with manual approach. It is observed that the use of SLRP-STC method can improve the planning of SLR.Keywords: Systematic Literature Review (SLR), planning phase; decision-making, Suffix Tree Clustering (STC), text mining (TM) INTRODUCTIONSystematic Literature Review (SLR), or Systematic Review (SR), provides a fair and reliable approach to identify, evaluate, and synthesize all primary studies, in attempt to address specific research questions [1]. SLR differs from ordinary review process in term of formal planning. Staples and Niazi [2] define a well-planned SLR as "independently replicable and of greater scientific value than ordinary literature reviews". For another, creating a comprehensive review plan is the key to the success of SLR as it drives the entire review methodology. Effective SLR planning takes into consideration all aspects of planning including identification of review need, specification of research questions, development, and validation of the review protocol.Performing SLR comprises three main phases: 1) planning the review, 2) conducting the review, and 3) reporting the review [5]. Particularly, planning phase is important to ensure the overall quality of SLR process. During the planning phase, reviewers are required to have adequate knowledge of the proposed topic area so they are likely to make a strategic plan for the SLR. For example, reviewers should confirm that there are likely to be sufficient relevant studies to make their SLRs worthwhile. They should also obtain the necessary information relevant to the studies, such as the number of independent studies that make review worthwhile, the research topics, the commonly used terms, and keywords. In this case, effective knowledge acquisition and discovery process can be useful and necessary in supporti...

Adoption of Machine Learning Techniques to Perform Secondary Studies: A Systematic Mapping Study for the Computer Science Field

“…In the following, we present the list of studies selected in this Systematic Mapping: S1 (Felizardo et al, 2010), S2 (Felizardo et al, 2011), S3 (Felizardo et al, 2012), S4 (Felizardo et al, 2014), S5 (Feng et al, 2017), S6 (Garcés et al, 2017), S7 (Malheiros et al, 2007), S8 (Mergel et al, 2015), S9 (Ouhbi et al, 2016), S10 (Piroi et al, 2015), S11 (Rizzo et al, 2017), S12 (Ros et al, 2017), S13 (Rúbio et al, 2016), S14 (Sellak et al, 2015), S15 (Tomassetti et al, 2011), S16 (Torres et al, 2013), S17 (Yu et al, 2018).…”

Automated Support for Searching and Selecting Evidence in Software Engineering: A Cross-domain Systematic Mapping

“…Variations of the Naïve Bayes classifier have been also explored [32], [34], [37] as well as Hybrid Feature Selection Method (HFSM) combined with other algorithms such as the hierarchical low-rank decomposition Blocked Adaptive Cross Approximation (BACA) [30] and the classical Rules7 [31]. Besides VTM, Suffix Tree Clustering and the optimization search algorithm Hill Climbing [39] are employed to automate the search for studies. Unlike other approaches, Ghafari et al [7] propose a federated search approach to automatic integrate search mechanisms across well-known SE DLs.…”

“…Regarding the specific tools that directly address automation of the activities of search and selection (see Table II), we detected two specific tools that provide automation support to SLR search activity [38], [39] and three specific tools to support the SLR study selection activity [27], [33], [50]. Three of the five identified tools uses VTM as automation technique [27], [38], [50].…”

“…Another highly adopted form of assessment for TC approaches is experiment considering data from published SLRs performed manually [18], [19], [28], [33], [36], [37], [43], [55], [62], [64], [65], [68], [71], [73]- [75], [81]. In these studies, the authors usually have two or more groups of participants to emulate the search or selection process using the proposed automated approach and compare their results against the manually performed search or selection process [16], [27], [38], [39].…”

“…2014 SLR.qub [38] Automated support the researcher by suggesting new terms for the string using VTM algorithms. 2015 SLRPSS [39] Unified search engine wrapper for the SLR DLs: IEEEXplore, ACM DL, the Web of Science, Science Direct, Scopus, and Google Scholar.…”