Abstract
dc:descriptionThis paper focuses on automatic requirements traceability and algorithms that automatically find recommendation links for requirements. The main objective of this paper is the evaluation of these algorithms and preparation of the method defining algorithms to be used in different cases. This paper presents and examines probabilistic, vector space and latent semantic indexing models of information retrieval and association rule mining using authors own implementations of these algorithms and one new feature of subsystem. After literature analysis and research, a method is defined on how automatic traceability between requirements and code should be carried out by companies and projects new to requirements management. Moreover, benefits, drawbacks and use cases are defined for algorithms of automatic requirements traceability. Conclusion of this paper states that automatic requirements traceability can be started without huge resources dedicated for it even if company or project are new to requirements management. However, in cases of informal and high level requirements it is crucial to choose algorithms wisely and thoroughly process text of artefacts. Probabilistic model is rarely a good choice in these cases event though it is most straight forward. Vector space model and latent semantic indexing are more suitable if carried out wisely.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Andžiulytė, Justė,
- Contributors dc:contributor
-
- Šurpicka, Ivona
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD23166622&prefLang=en_US