TU Darmstadt / ULB / TUbiblio

A Case Study of Simplified Syntax Trees as a Unified Data Structure for Recommender Systems in Software Engineering

Graf, Waldemar :
A Case Study of Simplified Syntax Trees as a Unified Data Structure for Recommender Systems in Software Engineering.
Technische Universität Darmstadt
[Bachelorarbeit], (2016)

Kurzbeschreibung (Abstract)

Recommender systems in software engineering(RSSE) often use intermediate representations(IR) to store data that is processed by the respective approach used. The choice of the data represents a vulnerability for the comparability in this research field. Often developers design new IRs for their approaches. This represents great effort and also has the risk of errors. This work contributes a unified data structure named Simple Syntax Tree(SST), which tries to address the issues mentioned for the RSSE research field. It is intended to combine the best properties of source code and bytecode. By providing the SST to RSSE developers the comparability of approaches that are used is highly increased. Additionally the effort to design and create an IR is reduced, allowing developers to focus on the approaches. Another important aspect of the introduced IR is, that it currently supports the C# and Java languages. The question whether the SST suffices as a unified data structure, will be discussed by listing the respective requirements and investigating if they are met. Additionally the SST is extendible to other languages and features in case that they are needed. In this thesis, the SST structure and the implementation used to create the SST is described in detail. The advantages and disadvantages of the SST to other IRs are also listed and compared. This is supposed to give evidence, if the SST can compete with them and where the strengths and shortcomings are.

Typ des Eintrags: Bachelorarbeit
Erschienen: 2016
Autor(en): Graf, Waldemar
Titel: A Case Study of Simplified Syntax Trees as a Unified Data Structure for Recommender Systems in Software Engineering
Sprache: Englisch
Kurzbeschreibung (Abstract):

Recommender systems in software engineering(RSSE) often use intermediate representations(IR) to store data that is processed by the respective approach used. The choice of the data represents a vulnerability for the comparability in this research field. Often developers design new IRs for their approaches. This represents great effort and also has the risk of errors. This work contributes a unified data structure named Simple Syntax Tree(SST), which tries to address the issues mentioned for the RSSE research field. It is intended to combine the best properties of source code and bytecode. By providing the SST to RSSE developers the comparability of approaches that are used is highly increased. Additionally the effort to design and create an IR is reduced, allowing developers to focus on the approaches. Another important aspect of the introduced IR is, that it currently supports the C# and Java languages. The question whether the SST suffices as a unified data structure, will be discussed by listing the respective requirements and investigating if they are met. Additionally the SST is extendible to other languages and features in case that they are needed. In this thesis, the SST structure and the implementation used to create the SST is described in detail. The advantages and disadvantages of the SST to other IRs are also listed and compared. This is supposed to give evidence, if the SST can compete with them and where the strengths and shortcomings are.

Fachbereich(e)/-gebiet(e): 20 Fachbereich Informatik
20 Fachbereich Informatik > Softwaretechnik
Hinterlegungsdatum: 18 Apr 2017 08:48
Gutachter / Prüfer: Mezini, Prof. Dr. Mira
Export:

Optionen (nur für Redakteure)

Eintrag anzeigen Eintrag anzeigen