A knowledge-workbench-for-software-development

A Knowledge Workbench for Software Development
Dimitris Panagiotou
(National Technical University of Athens, Athens, Greece
dpana@mail.ntua.gr)
Gregoris Mentzas
(National Technical University of Athens, Athens, Greece
gmentzas@mail.ntua.gr)
Abstract: Modern software development is highly knowledge intensive; it requires that
software developers create and share new knowledge during their daily work. However, current
software development environments are “syntactic”, i.e. they do not facilitate understanding the
semantics of software artefacts and hence cannot fully support the knowledge-driven activities
of developers. In this paper we present KnowBench, a knowledge workbench environment
which focuses on the software development domain and strives to address these problems.
KnowBench aims at providing software developers such a tool to ease their daily work and
facilitate the articulation and visualization of software artefacts, concept-based source code
documentation and related problem solving. Building a knowledge base with software artefacts
by using the KnowBench system can then be exploited by semantic search engines or P2P
metadata infrastructures in order to foster the dissemination of software development
knowledge and facilitate cooperation among software developers.
Keywords: Software Development, Knowledge Workbench, Semantic Annotation, Semantic
Wiki
Categories: M.0, M.3, M.4
1 Introduction
Modern software development consists of typical knowledge intensive tasks, in the
sense that it requires that software developers create and share new knowledge during
their daily work. Although most software developers use modern state of the art tools,
they still struggle with the use of technologies that are “syntactic”, i.e. they use tools
that do not facilitate the understanding of the concepts of the software artefacts they
are managing (e.g. source code).
Flexible ways of solving problems are necessary when a developer is frustrated
investigating source code that he has never seen before (i.e. when extending a third
party’s software system) and is not capable of understanding its rationale.
Additionally, there are many situations that find a developer seeking source code that
is already developed by others. He might not be aware of its existence, or even if he
is, he is not able to find it effectively [Bauer, 06], [Thaddeus, 06].
Recent trends in data integration and combination have led to the Semantic Web
vision1
. The Semantic Web strives to provide integration among information scattered
1
http://www.hpl.hp.com/semweb/sw-vision.htm

across different locations in a uniform fashion. It defines a “common language”
between these information resources.
Additionally, there is a growing interest in the last few years on exploiting
Semantic Web technologies in software engineering. Since 2005 the Semantic Web
Enabled Software Engineering (SWESE) conference takes place every year with
promising results. The Semantic Web Best Practice and Deployment Working Group
(SWBPD) in W3C included a Software Engineering Task Force (SETF) to investigate
potential benefits of applying semantics in software engineering processes. As noted
by SETF [Primer, 06], advantages of applying Semantic Web technologies to
software engineering include reusability and extensibility of data models,
improvements in data quality, enhanced discovery, and automated execution of
workflows.
In this paper we propose a new system – code-named KnowBench for knowledge
workbench – which is the result of exploiting Semantic Web technologies in the
context of a well known software development IDE such as Eclipse2
, in order to
provide the basis for better collaboration in the software development domain and
strive against the aforementioned problems. KnowBench aims at providing software
developers such a tool to ease their daily work and facilitate source code
documentation and problem solving.
The rest of the paper is organized as follows. In the next section, we describe
what motivated us in developing the KnowBench system. Afterwards, we describe its
functionality. Thereafter, we give the results of its preliminary evaluation and we
discuss the benefits of using the system. Finally, we conclude the paper indicating our
thoughts for future work.
2 Motivation
Modern IDEs such as Eclipse provide means for documenting and annotating source
code, e.g. JavaDoc and Java annotations. However, this kind of documentation is
unstructured and the means to retrieve it are limited.
In most cases, software developers when in need of specific functionalities find
themselves seeking in forums, web sites or web search engines using keyword search
for an API capable of delivering these functionalities. Additionally, software
developers often use dedicated forums and mailing lists to report on bugs and issues
they have with specific source code fragments and have to wait for answers from
experts. This switching of environments while developing code is obtrusive and
requires a great amount of effort in order to maintain scattered resources of software
development related knowledge [Gall, 08].
We argue that a more formal description of the aforementioned knowledge could
be beneficiary for the software developers and the means to articulate and visualize
this knowledge should be unobtrusive i.e. in the IDE itself in order to provide unified
access to it. On top of that, knowledge about source code or software components
should be described in a common way in order to avoid ambiguities when this
knowledge is retrieved and used by others.
2
http://www.eclipse.org/

Documentation of software, bug fixing and software related knowledge transfer
between employees of a large software house can be facilitated by the use of
ontologies to sufficiently capture and describe software artefacts in a coherent way.
Furthermore, ontologies are ideal for interlinking software artefacts and the resulting
semantic relationships can be exploited in order to maximize productivity when
seeking related knowledge of a specific software artefact.
Towards this end we have been motivated to design and implement an
environment – namely KnowBench – that offers the opportunity to articulate and
visualize software artefacts driven by an ontology backbone. The created artefacts
formulating an organization’s software development knowledge base can be exploited
by semantic search engines and P2P metadata infrastructures to enable knowledge
sharing and reuse in order to boost the collaboration among the employees and
speedup time-consuming procedures when facing problems with code or searching for
APIs or components to be reused.
3 KnowBench
The KnowBench is designed and implemented in order to assist software
development work inside the Eclipse IDE exploiting the power of semantic-based
technologies (e.g. manual/semi-automatic semantic annotation, software development
semantic wiki).
KnowBench exposes functionality that can be used for articulating and
visualizing formal descriptions of software development related knowledge in a
flexible and lightweight manner. This knowledge can be then retrieved and used in a
productive manner by integrating the KnowBench with a semantic search engine and
a P2P metadata infrastructure such as GridVine [Mauroux, 07]. This integration will
assist the collaboration and the formation of teams of software developers who can
benefit from each others’ knowledge about specific problems or the way to use
specific source code while developing software systems. We have already integrated
and tested KnowBench with a semantic search engine and GridVine in the context of
the TEAM project [Team, 08]. However, the description of these systems is out of the
scope of this paper. A demo of the TEAM system is available online3
.
3.1 Ontologies
We have deployed a set of software development ontologies to the KnowBench
system in order to describe and capture knowledge related to software artefacts. The
system architecture allows for the extension or even the use of different software
development ontologies. The set of the deployed ontologies constitutes of three
separate ontologies.
The artefact ontology describes different types of knowledge artefacts such as the
structure of the project source code, reusable components, software documentation,
knowledge already existing in some tools, etc. These knowledge artefacts typically
contain knowledge that is rich in both structural and semantic information. Providing
a uniform ontological representation for various knowledge artefacts enables us to
3
http://www.team-project.eu/demo.php

utilize semantic information conveyed by these artefacts and to establish their
traceability links at the semantic level.
An artefact is an object that conveys or holds usable representations of different
types of software knowledge. It is modelled through the top level class Artefact.
There are some common properties for all types of artefacts such as:
 hasID – which identifies an artefact in a unique manner. Recommended best
practice is to identify the artefacts by means of a string conforming to a
formal identification system;
 hasTitle – which defines a title of an artefact (e.g. class name or document
title);
 hasDescription – which describes what an artefact is about;
 createdDate – which models date of creation of an artefact;
 usefulness – which models how frequently an artefact is used as a useful
solution.
The problem/solution ontology models the problems occurring during the
software development process as well as how these problems can be resolved. This
ontology is essential to the KnowBench system, as source code can be documented
when a certain problem is met and the respective solution to it is described.
A problem is an obstacle which makes it difficult to achieve a desired goal,
objective or purpose. It is modeled through the class Problem and its properties:
 hasID – a problem is identified by its identifier;
 hasTitle – a summary of a problem;
 hasDescription – a textual description of a problem;
 hasSeverity – it models how severe the problem is;
 hasSolution – every problem asks for a solution. This is the inverse property
for the property isRelatedToProblem defined for the class Solution;
 hasSimilarProblem – it models similar problems.
A solution is a statement that solves a problem or explains how to solve the
problem. It is represented through the Solution class and its properties:
 hasID – an unambiguous reference to a solution;
 isRelatedToProblem – a solution is defined as the means to solve a problem;
 hasCreationDate – date of creation of a solution;
 hasDescription – a textual description of a solution;
 hasPrerequirement – a solution requires a level of expertise in order to be
used;
 hasComment – a solution can be commented by the people that used it;
 suggestsArtefact – a solution recommends using an artefact in order to
resolve a problem the solution is defined for;
 isAbout – a solution can be annotated with the domain entities (see below);
 isRelatedToSEEntity – a solution can be annotated with the general
knowledge about software engineering domain (see below).
The annotation ontology describes general software development terminology as
well as domain specific knowledge. This ontology provides a unified vocabulary that
ensures unambiguous communication within a heterogeneous community. This
vocabulary can be used for the annotation of the knowledge artefacts. We distinguish
two different types of annotations: (a) domain annotation – software providers in

different domains should classify their software using a common vocabulary for each
domain. Common vocabulary is important because it makes both the users and
providers to talk to each other in the same language and (b) software engineering
annotation – general knowledge about the software domain including software
development methodologies, design patterns, programming languages, etc.
By supporting different types of annotation it is possible to consider information
about several different aspects of the artefacts. We note here that the vocabulary will
be used for the annotation of the artefacts, e.g. code, knowledge documents, etc. as
well as expertise of the users and solutions of problems. These annotations will be
used to establish the links between these different artefacts.
3.2 Software Development Semantic Wiki
KnowBench utilizes the SoWiSE2 system which is the successor of SoWiSE
[Panagiotou, 08] in order to assist software developers in the articulation and
navigation of software development related knowledge. SoWiSE is a Semantic Wiki
for Software Development and provides common functionalities with a conventional
wiki with semantic capabilities. It provides software knowledge articulation support
with the usage of ready to use templates. On the other hand, SoWiSE2 uses a
lightweight and flexible editor with auto-completion and popup support to eliminate
the need of templates which sometimes are cumbersome to use. Browsing through
knowledge is done like surfing through a conventional wiki using the semantic links
between different knowledge artefacts. This browser is available inside the Eclipse
IDE so that the software developer does not have to switch to another external
browser.
SoWiSE2 provides assistance to software developers in accomplishing their tasks
in a more flexible fashion and shortens their total effort in terms of time. In order to
achieve this goal, we approach software development documentation and related
problem solving by combining lightweight yet powerful wiki technologies with
Semantic Web standards.
An overview of the provided functionality is given below:
 Validation of the user input when incorrect values for selected concepts,
properties, related instances or property values are introduced. The semantic
wiki informs the user about the wrong input value and assists him/her in the
way to track down the problem and resolve the issue.
 Auto-complete facilities provide assistance to the user for selecting the right
concept, properties and related instances of an object property as a value.
 Intuitiveness – the software developer is able to understand what is required
as input using the auto-complete facilities of the Semantic Wiki.
 Syntax colouring – the software developer is assisted with syntax colouring
to easily determine what kind of information is recognized by the wiki
engine in order to avoid mistyping.
 Multipage editor consisting of the semantic wiki editor and the semantic wiki
browser which facilitates navigation of the knowledge base in an HTML-
fashion inside the Eclipse environment.
 Navigation through the semantic wiki is enabled via following an instance’s
semantic links to other related instances.

 On the fly synchronization with the knowledge base. SoWiSE2 keeps no files
anywhere and all information is persisted directly to the knowledge base.
Figure 1 depicts the user interface of SoWiSE2.
Figure 1: SoWiSE2 browser and editor
3.3 Manual Semantic Annotation of Source Code
An important aspect of the KnowBench is the ability to annotate semantically source
code. We have extended the standard Eclipse JDT editor to add this possibility. The
software developer is able to annotate source code with semantic annotation tags that
are available or define new tags and extend the used annotation ontology.
The manual semantic annotation of source code provides granularity regarding
the respective source code fragment to be annotated. This granularity level is
restricted by the underlying Eclipse platform itself as the IJavaElement interface is
exploited to map between source code fragments and metadata. This limits the
selected source code fragments to be annotated in the nearest Java elements that
surround the source code fragment at hand (e.g. package/class/method/variable etc.).
Additionally, by using IJavaElement the semantic annotation mechanism is capable of
managing annotations of several objects, such as Java packages, classes, methods, etc.
3.4 Semi-automatic Semantic Annotation of Source Code
This kind of annotation is supervised by the user and results in new concepts and
sometimes in their classification inside the annotation ontology hierarchy tree. These
new or existing concepts of the annotation ontology are then used to annotate

semantically the source code corpora. This procedure assists and shortens the time
consuming task of manually annotating source code.
The need for assisting the software developer in annotating semantically source
code (either one single file of code or a batch of files residing in a directory) is
evident. The notion of semi-automatic comprises a user-controlled process with the
system acting as a mediator (proposing semantic annotation concepts/taxonomies of
concepts) and the user having the final decision of acceptance or rejection or revision
and acceptance of the system's propositions. Thus, the time consuming task of
manually annotating source code can be significantly decreased.
In order to achieve this goal in KnowBench we had to elaborate on ontology
learning [Maedche, 04] and information extraction techniques [Cunningham, 02].
Ontology learning is needed in order to extract ontology concepts from source code
corpora that can be used for annotating it. On the other hand, information extraction is
needed in order to instantiate the annotation ontology with individuals (like in manual
semantic annotation), either of newly proposed and created or existing ontology
concepts. A framework exploiting both technologies is Text2Onto [Cimiano, 04]
which we adopted in order to implement the desired functionality.
The architecture of Text2Onto is based on the Probabilistic Ontology Model
(POM) which stores the results of the different ontology learning algorithms. A POM
as used by Text2Onto is a collection of instantiated modelling primitives which are
independent of a concrete ontology representation language.
KnowBench takes advantage of two modelling primitives, i.e. concepts and
concept inheritance. In the respective modelling primitives KnowBench employs
different Text2Onto algorithms in order to calculate the probability for an instantiated
modelling primitive. In particular the following measures are calculated:
Concepts are identified using the following algorithms: Relative Term Frequency
(RTF), Term Frequency Inverted Document Frequency (TFIDF), Entropy and the C-
value/NC-value method in [Frantzi, 98]. For each term, the values of these measures
are normalized into the interval [0..1], after taking the average probability of the
algorithms output combination which is used as the corresponding probability in the
POM.
Regarding concept inheritance, KnowBench combines Text2Onto algorithms
such as matching Hearst patterns [Hearst, 92] in the corpus and applying linguistic
heuristics mentioned in [Velardi, 05].
The KnowBench semi-automatic semantic annotation editor is depicted in Figure
2. After processing a source code corpus, the developer can supervise and change the
proposed annotations and their position in the concept hierarchy as per his/her needs
using this editor. He/she is able to filter the results dynamically by dragging the
threshold slider, which means that if a given proposed annotation concept has a lower
confidence value than the selected threshold then this is not included in the list, which
is refreshed dynamically. The confidence percentage is the calculated probability of a
concept to be precise and it reflects the POM probability of a term. Additionally, the
developer can rename the proposed concepts and change their position in the concept
hierarchy (by choosing their parent concept in the annotation ontology tree).
Changing the position of a concept is automatically reflected on the annotation type
(Domain or Software Engineering).

Figure 2: Semi-automatic semantic annotation editor
4 Usage Scenario
Dave is writing an XML editor as a plug-in for Eclipse. He uses the semi-automatic
semantic annotation mechanism to annotate the source code. He reviews the
suggestions provided by the system (see Figure 3) and submits them to the system.
Figure 3: Annotation suggestions for source code of an XML editor
He also describes some guidelines in the wiki about “validating and character
encoding XML code” as in Figure 4.
John is a new Eclipse developer and intends to write an XML editor. He is
connected to the P2P network and performs a structured query to find knowledge or
source code that is related to “XML editor” as in Figure 5.
He retrieves the results as in Figure 6 and now he is able to reuse existing source
code and take into consideration the guidelines documented by Dave.

Figure 4: Wiki page describing guidelines about “validating and character encoding
XML code”
Figure 5: Structured P2P query for finding knowledge that is related to “XML
editor”
Figure 6: Search results of the query sent as in Figure 5
5 Preliminary Evaluation
We have conducted a preliminary evaluation of KnowBench in small teams of
software developers in a total of 9 persons in the following organizations: (1) a
Brussels based company specializing in the field of Information and Communication
Technology (ICT) services (Intrasoft International S.A. – 2 developers), (2) a leading
hungarian association dealing with open source software at corporate level (Linux
Industrial Association – 2 developers), (3) an italian company which operates in the
Information Technology market, focusing on business applications (TXT e-Solutions
– 2 developers) and (4) the corporate research laboratory of the Thales group – a

global electronics company serving Aerospace, Defence, and Information Technology
markets worldwide (Thales Research & Technology – 3 developers). A more formal
evaluation of the system is currently underway and the results are expected in the next
few months.
Overall, the KnowBench system has been well appreciated by the software
developers. It is a simple, user-friendly and intuitive system aiming to improve their
productivity. In that respect, the provided functionalities seem to integrate well with
the Eclipse IDE without encumbering the developer with time-consuming functions.
In fact, these functionalities demonstrate a good value proposition for the developers’
work.
The initial effort needed to start working with the system is minimal;
nevertheless, it seems that some of its functions convey deeper logic which is not
immediately capturable by the end users. The graphical user interface could be
improved further in order to reveal this logic, while some help functions could also be
of great value on this. Users were also satisfied by the level of “contextualization”
offered by the system (meaning helping the developer know where he/she is at every
moment). Finally, the performance of the KnowBench system i.e. response time left
the developers with very good impressions.
More specifically, annotations are a powerful tool in managing knowledge and
the KnowBench implementation seems to be very efficient. Developers were rather
comfortable with the use of annotations and could use them in an intuitive manner i.e.
name annotations freely. Additionally, semi-automatically creating annotations was
even more appreciated and greatly assisted them in speeding up the error-prone and
cumbersome annotation process. However, they did not seem to understand their real
value and thus their deeper connection with the inner logic of the system.
The developers showed great comfort in using the SoWiSE2 wiki of the
KnowBench. Instantiation takes place quickly and the types of information that are
needed for the creation of a wiki page are clear, while interactive mechanisms i.e.
pop-ups make the creation a very simple task.
6 Discussion
There are emerging efforts in research towards exploiting Semantic Web technologies
in software engineering. For example, SemIDE [Bauer, 06] is a framework which is
based on meta-models in order to describe semantically software engineering
processes. OSEE [Thaddeus, 06] is a tool which aids the construction of a knowledge
base which captures software related artefacts. SeRiDA [Athanasiadis, 07] combines
object-oriented programming and relational databases, with semantic models.
With KnowBench integrated into an IDE, it is easier for software developers to
create new knowledge. The advantage of recording the solution to difficult or
frequently asked questions/problems is that developers are recording their explicit
knowledge about particular problems, as well as tacit knowledge that they have
internalized, while creating software development knowledge artefacts that can be
organized, managed, and reused.
Wikis can significantly help developers fill this knowledge management need.
KnowBench provides this kind of wiki-based support for knowledge provision.
SoWiSE2 offers innovative features especially in software development but is also a

contemporary semantic wiki in the general sense since it provides many features
common in most of the state-of-the-art semantic wikis [Maalej, 08].
KnowBench is seamlessly integrated into the Eclipse environment and thus gives
software developers the incentives to develop or explore knowledge without
bothering to open an external tool or web browser to accomplish this task which has
the advantage of leaving the user in his/her current work context and not distracting
him/her from the rest of his/her tasks. For example, he/she can develop source code
and at the same time document problems that he/she encounters or provide guidelines
and workarounds for a specific problem in source code that is commonly known to
his/her organization by simply choosing to open the SoWiSE2 editor next to the
source code editor.
7 Conclusion
In this paper we presented the KnowBench system, a knowledge workbench which is
integrated in the Eclipse IDE based on Semantic Web technologies, in order to assist
software developers in articulating and visualizing software development knowledge.
The resulting knowledge base can be exploited by semantic search engines or P2P
metadata infrastructures in order to foster better and more flexible collaboration
among software developers scattered across the globe.
We aim at providing support to the software developer for documenting code,
finding the best solution to a problem or the way to move forward in critical decisions
in order to save time during the heavy task of software development. Thus, in
KnowBench we provide an easy to use semantic wiki.
Since manual creating knowledge may be very time-consuming and error-prone
activity, the KnowBench system takes advantage of ontology learning and
information extraction techniques to propose annotations. Both manual and semi-
automatic semantic annotation of source code is facilitated by automated generation
of corresponding metadata that assist also in describing and mapping Eclipse
resources to the knowledge base.
We are planning to enhance the semi-automatic semantic annotation mechanism
in order to derive more precise annotations by introducing JAPE grammar rules
[Cunningham, 02] of the GATE system that will be customized for source code.
Furthermore, we want to further extend the manual semantic annotation in order to
add the capability of annotating arbitrary lines of code regardless of the surrounding
Java elements. However, this is a hard to solve and non-trivial issue.
Acknowledgements
This work was partly supported by the TEAM project, which is funded by the EU-IST
program under grant FP6-35111. The authors are responsible for the content of this
publication.

References
[Athanasiadis, 07] Athanasiadis, I.N., Villa, F., Rizzoli, A.E.: Enabling knowledge-based
software engineering through semantic-object-relational mappings, In Proc. 3rd International
Workshop on Semantic Web Enabled Software Engineering (2007)
[Bauer, 06] Bauer, B., Roser, S.: Semantic-enabled Software Engineering and Development, In
Proc. Hochberger, C., Liskowsky, R. (eds.) INFORMATIK 2006 - Informatik für Menschen.
LNI, vol. P-94, pp. 293-296, Bonner Köllen Verlag (2006)
[Cimiano, 04] Cimiano, P., Pivk, A., Schmidt, L.T., Staab, S.: Learning taxonomic relations
from heterogeneous sources, In Proc. ECAI 2004 Ontology Learning and Population Workshop
(2004)
[Cunningham, 02] Cunningham, H., Maynard, D., Bontcheva, K., Tablan, V.: GATE: A
framework and graphical development environment for robust NLP tools and applications, In
Proc. 40th Annual Meeting of the ACL (2002)
[Frantzi, 98] Frantzi, K., Ananiadou, S., Tsuji, J.: The c-value/nc-value method of automatic
recognition for multi-word terms, In Proc. ECDL, pp. 585-604 (1998)
[Gall, 08] Gall, H.C., Reif, G.: Semantic Web Technologies in Software Engineering, In Proc.
Schafer, W., Dwyer, M.B., Gruhn, V. (eds.) 30th International Conference on Software
Engineering. Leipzig, Germany, May 10-18, 2008, ACM (2008)
[Hearst, 92] Hearst, M.: Automatic acquisition of hyponyms from large text corpora, In Proc.
14th International Conference on Computational Linguistics, pp. 539-545 (1992)
[Maalej, 08] Maalej, M., Panagiotou, D., Happel, H.J.: Towards Effective Management of
Software Knowledge Exploiting the Semantic Wiki Paradigm, In Proc. Software Engineering
2008, pp. 183-197 (2008)
[Maedche, 04] Maedche, A., Staab, S.: Ontology learning, In Proc. Staab, S., Studer, R. (eds.),
Handbook on Ontologies, pp. 173-189. Springer (2004)
[Mauroux, 07] Mauroux, P.C., Agarwal, S., Aberer, K.: GridVine: An Infrastructure for Peer
Information Managment, IEEE Internet Computing, vol. 11, no. 5, pp. 36-44 (2007)
[Panagiotou, 08] Panagiotou, D., Mentzas, G.: Exploiting Semantics in Collaborative Software
Development Tasks, In Proc. Vivrou, M., Nakamura, T. (eds.) 8th Joint Conference on
Knowledge-Based Software Engineering (JCKBSE). Frontiers in Artificial Intelligence and
Applications, vol. 18, pp. 385-394, IOS Press (2008)
[Primer, 06] A Semantic Web Primer for Object-Oriented Software Developers, (2006),
http://www.w3.org/TR/sw-oosd-primer/
[Team, 08] TEAM - Tightening knowledge sharing in distributed software communities by
applying semantic technologies, (2008), http://www.team-project.eu/
[Thaddeus, 06] Thaddeus, S., Kasmir, R.S.V.: A Semantic Web Tool for Knowledge-based
Software Engineering, In Proc. 2nd International Workshop on Semantic Web Enabled
Software Engineering (SWESE 2006), Athens, G.A., USA, Springer (2006)
[Velardi, 05] Velardi, P., Navigli,, R., Cuchiarelli, A., Neri, F.: Evaluation of ontolearn, a
methodology for automatic population of domain ontologies, In Proc. Buitelaar, P., Cimiano,
P., Magnini, B. (eds.). Ontology Learning from Text: Methods, Applications and Evaluation.
IOS Press (2005)

A knowledge-workbench-for-software-development

More Related Content

Similar to A knowledge-workbench-for-software-development (20)

Recently uploaded (20)

A knowledge-workbench-for-software-development