Combining Machine Learning and Logical Reasoning to Improve Requirements Traceability Recovery

Abstract

Maintaining traceability links of software systems is a crucial task for software management and development. Unfortunately, dealing with traceability links are typically taken as afterthought due to time pressure. Some studies attempt to use information retrieval-based methods to automate this task, but they only concentrate on calculating the textual similarity between various software artifacts and do not take into account the properties of such artifacts. In this paper, we propose a novel traceability link recovery approach, which comprehensively measures the similarity between use cases and source code by exploring their particular properties. To this end, we leverage and combine machine learning and logical reasoning techniques. On the one hand, our method extracts features by considering the semantics of the use cases and source code, and uses a classification algorithm to train the classifier. On the other hand, we utilize the relationships between artifacts and define a series of rules to recover traceability links. In particular, we not only leverage source code’s structural information, but also take into account the interrelationships between use cases. We have conducted a series of experiments on multiple datasets to evaluate our approach against existing approaches, the results of which show that our approach is substantially better than other methods.

Clustering-based Automated Requirement Trace Retrieval

“…In the paper [24] titled "Combining Machine Learning and Logical Reasoning to Improve Requirements Traceability Recovery", the authors proposed a novel traceability link recovery approach that measures the similarity between requirements and the source code by exploring their features. www.ijacsa.thesai.org They combined machine learning and logical reasoning models and conducted a series of experiments on four datasets to evaluate the performance of their method against existing approaches.…”

“…Comparing the proposed approach with two experiments, namely TraceLab and IR as shown in Tables IV (A) to (C). Tables V (A) to (C) show the comparison results of the proposed method and the two studies (i.e., studies [19] and [24]).…”

“…The extent of improvement differed from one dataset to another depending on the dataset type. We used accuracy as an additional measure, whereas TraceLab and Information retrieval R and two studies [19] and [24] did not. The proposed approach achieved highest results using accuracy with the three datasets (i.e., eTour = 0.91, SMOS = 0.66, and eANSI = 0.60).…”

“…. eTour result comparison: clustering and studies [19] and [24] (B). SMOS result comparison: clustering and studies [19] and [24] (C).…”

“…eTour result comparison: clustering and studies [19] and [24] (B). SMOS result comparison: clustering and studies [19] and [24] (C). eANCI result comparison: clustering and studies [19] and [24] Finally, the authors will present all the experimental results obtained from K-means++, hierarchical, and DBSCAN as shown in (Tables VI-XVII).…”