VOL 2 .
NO 4
e-ISSN : 2549-9904
ISSN : 2549-9610
INTERNATIONAL JOURNAL ON INFORMATICS VISUALIZATION
Developing Research Questions in Natural Language Processing and Software Engineering Nakul Sharma#.
Prasanth Yalla# # Koneru Laxminah Education Foundation.
India E-mail: nakul777@gmail.
com, prasanthyalla@gmail.
AbstractAi This paper endeavors to develop newer medium of developing research questions by keeping in view both fields of SE and NLP in proper perspectives.
An overview of the current state of art research in SE and NLP is presented.
This is done by referring to the SE Body of Knowledge (SEBOK).
Analogues to SEBOK, there are no separate Body of Knowledge available for the NLP/Computational Linguistics (CL).
Hence whatever falls within the category of NLP/CL was considered in framing the research categories from the NLP/CL side.
The paper concludes with future scope of the research presented.
KeywordsAi Software Engineering.
Natural Language Processing.
Computational linguistics.
Engineering gives an insight to the sub-areas in which research can be conducted.
INTRODUCTION
A research question holds the key to successful research.
Software Engineering along with Natural Language Processing are emerging areas in Computing environment.
They can be applied to various situations and contexts in which software or any textual content are developed and Software Engineering tasks, methods, processes, formal methods etc.
Natural Language Processing also consists of various tasks and methods on which Software Engineering can be applied.
The methods can be classical approach to NLP or Empirical approach to NLP.
II.
Table-I summarizes the existing work in regard to the sub-areas in Software Engineering .
TABLE I
TABLE I.
SOFTWARE ENGINEERING DOMAINS FROM** .
Parameter Software Process In the current work, the focus is providing better perspectives in framing the most critical aspect of research, research questions.
In this work, the endeavor is to provide a framework for generating research questions by putting together both the areas of research into perspective.
This is given in Equation1.
(NLP)sub-area***(SE)sub-area SOFTWARE ENGINEERING DOMAINS Process Models .
The concepts in both the areas can be both qualitative as well as quantitative.
Equation-1 can be used in different context in developing the research questions.
The * sign can represent a AND sign.
The AND sign implies that both research areas must be taken into consideration while doing A generic methodology is generation of research questions can be evolved.
A cursory reading of Software Agile Development Requirement Software Engineering Areas CMMI Process Process Patterns Personal & TSP Process Process Technology The most traditional model, waterfall model additive process models includes RAD and incremental models.
There are some process models which increase which are Evolutionary in nature:-Prototyping model -Concurrent model -Spiral model Specialized process models -Component Based -Formal methods -Aspect-oriented software development Basic Methodologies in Agile development Extreme Programming Adaptive Software Development (ASD) Dynamic Systems Development Method Scrum Feature Driven Development Tasks Engineering Analysis Model Architectural Design Product Metrics Quality Management Inception Elicitation Elaboration Negotiation Specification Validation Requirement Management Requirement Analysis Data Modeling Concepts Analysis Modeling using scenarios Modeling using Data Flow Class-Based modeling This includes software as a architecture.
Designing the Data, its flow within the The styles and patterns involving Some alternative ways of designing the architecture.
Checking how the data gets flow from the software architecture.
Software Metrics for each models in SDLC Software Configuration Management Management within Software Engineering Software Engineering Process Software Quality Assurance Software Reviews Formal Technical Review Software Engineering Models and Methods Software Quality i.
SOFTWARE ENGINEERING DOMAINS ACCORDING TO
SEBOK
According the Software Engineering Body of Knowledge, there are following research areas in Software Engineering .
:- Software Engineering Economics TABLE II SOFTWARE ENGINEERING DOMAINS FROM** .
Parameter Software Requirements Software Design Development of Software Software Testing Software Maintenance Software Engineering Areas Requirements Basics Requirement Process Requirement induction Requirement Analysis Development of SRS Validation of requirements through test case generation.
Design of Software Basics Architecture and structure of software UI Design The Design of Software Quality Analysis and its Evaluation Notations for Software Design Software Design Strategies and Methods Tools used in designing the software Managing Construction Construction Techniques Tools employed in development of Software Testing Fundamentals The different levels within testing Different Techniques in undertaking Test-Related Measures The Process of Testing Tools for testing the software.
Software Maintenance Basics Main points in Software Maintenance Maintenance Process Techniques for Maintenance Identification of different configurations for softwares being developed.
Controlling the software configuration.
Software Configuration Status Accounting How software is delivered and released to the clients.
Software Configuration Auditing Software Configuration Management Tools Initiation and Scope Definition Software Project Planning Software Project Enactment Review and Evaluation Closure Software Engineering Measurement Software Engineering Management Tools Software Life Cycles Software Process Assessment and Improvement Software Measurement Software Engineering Process Tools Types of Models 2.
Analysis of Models Software Engineering Methods Software Quality Management Processes Software Quality Tools Life Cycle Economics Risk and Uncertainty Economic Analysis Methods IV.
SOFTWARE ENGINEERING DOMAINS ACCORDING TO
SEBOK
There are two types of NLP approaches are classified as .
:1.
Classical Approaches to NLP Empirical and statistical approach to NLP Classical Approaches to NLP The classical approach to NLP consists of following major sub-areas:1.
Text Processing Lexical Analysis Syntactic Parsing Semantic Analysis Natural Language Generation Empirical and Statistical approach to NLP Corpus creation Treebank annotation SMT POS Tagging Statistical Parsing Multiword Expressions Word Sense Disambiguation METHODOLOGY FOR GENERATING RESEARCH QUESTION In section-II,i,IV different sub-areas within Software Engineering domain as well as Natural Language Processing are shown.
Using the equation-1, different sub domains of SE and NLP can be researched for.
As this involves more than one research area, interdisciplinary research methodology is presented named Research Question Interdisciplinary (RQI).
The technique RQI involves following steps:1.
Identify if the research is to be done more in SE domain or NLP domain.
1 If the research is done in SE domain then just take some few for research 2 If the research is done in NLP domain then include some sub-areas of SE for research Repeat Stop after getting necessary research papers.
This can hence help in creating a taxonomy of keywords.
The keywords can be put in Equation-I and can be used to get relevant research papers.
There can be 1:m relationship being applied or m:1 relation being applied as given in equation-1.
So one or more parameter of SE can be applied to NLP.
Reverse is also true.
One or more parameter of NLP can be applied to SE.
Generating Research Questions Generating Research directions Generating Research Projects As SE and NLP itself contain lot of sub-areas, it is hence possible widen to scope of research and resulting in better products being developed.
VII.
CONCLUSION
This paper provides a research direction for developing research questions by combining SE and NLP areas of The summarizes the essential portions of SE and NLP.
The authors provide the methodology for generating research questions.
Essentially, the research methodology presented can be applied to generate numerous research
questions and hence projects which will increase innovation in the field of interdisciplinary research
REFERENCES