The purpose of this report is to consider and analyze similarities and differences between the four scientific articles with reference to S rensen s article This is not an Article . For this report we have chosen the topic in the area of computer science.
Article, analysis, Information System development (ISD), research reporting, software development practice, Emotional Learning (EL), Cognitive Tutor System (CTS), Face, Facial Recognition, Object Detection, Face Recognition, Face Processing, Face Recognition Algorithms, Face Detection, software testing, software development process
This report contains the result of analyzing and comparing similarities and differences of four chosen articles in terms of abstracts, introduction, method, result, discussion and conclusion according to the S rensen s article .
Based on S rensen s article analysis of above elements in articles has to be able to answer following questions:
* What is the problem domain?
* What is the problem?
* What is the research approach?
* What have others done?
* What are results?
The four scientific articles that we have opted are:
Article1: The NIPO GRID A Construct for systems development practices in organization .
Article2: Implementation of Emotional Learning for Cognitive Tutoring Agents .
Article3: Best Practices for the Organizational Implementation of Software Testing .
Article4: Face recognition for smart environment .
In the following sections differences and similarities of the above scientific articles are summarized according to the essential questions.
1.1 Abstracts, Contexts
In general, abstracts summarize the article contents of purpose, method, results and conclusion .
In the first article , the research is written in the context of software development practices and is shaped on the basis of the question how and why actual practices are shaped in professional system and software development organizations. The abstract provides concise, clear and interesting overview of what the reader will gain by reading the article. Plus, Along with following abstract components , writers avoided using references, jargons and acronyms in the abstract.
In the second article  the short abstract contains a reference at the end of abstract and an acronym along with complete description has been mentioned. The Context is in learning mechanism and proposed mechanism for improving Cognitive Tutoring System (CTS) using Emotional Learning (EL). The abstract determined the goal of the article, which is about increasing agent s autonomy in a changeable environment.
The third article s  abstract, summarized the result of a project with regional companies in software testing. The article describes four main recommendations considering the organizational implementation and strategic adjustment of testing. It introduces a framework to classify recommendations for testing.
In the fourth article , the authors discuss about face recognition technology in smart environment as assistance, for helping humans to be successful in absorbing wide consumer approval
On the whole, abstract in the first article  provides concise, clear and interesting overview of what the reader will gain by reading the article while in the second article  writers used a reference and an acronym at the end of the abstract, besides the point the problem is not directly mentioned. In the third article  abstract introduces the research filed and also identifies the main problem with the filed. It also avoids using reference and acronym. In the fourth article , abstract was not mentioned as a separate subtitle, however the first paragraph has the same characteristics of abstract and gives the reader a complete and quick identification of the basic content of the article.
1.2 Problem Domains and Motivations
Presenting the problem domain helps focusing the attention of the reader and frames the research problem .
The first article  presents intermediate results from a study of issues, which influence shaping of information systems development (ISD) practices in systems and software development organizations. It describes a construct for studying the behavior of organizations in using Software Development Method practices in the process of product development. While the paper is part of an ongoing project of a company it studies the mentioned behavior from two dimensions, the attitude and execution as rules. The report itself uses terms "intended practices" vs. "enacted practices".
The second article  is a research-base paper; it searches within learning mechanisms area with focusing on emotional learning and defines another related approach that improves the existing CTS. The main purpose of this scientific article is to equip the CTS by an emotional learning mechanism to work in changeable environment like human being.
The problem domain of the third article  is how to improve software quality .it analysis the software testing as a part of the software development process of the companies to extract their problems in software development and to use their best practices in software testing.
The fourth article  argues about what Technology is the best for people identification and discuss about face recognition technology, how it works, the problems, current technologies and future development.
1.3 Research Questions
One of the main factors should be considered in evaluating document, is how the research questions have been shaped and in the following how suitable the methodology is selected and whether the selected research method has led to finding the answer to the research questions or not.
In the first article  there are two questions raised and all answered based on the observations from the real companies:
* How and why actual ISD practices are shaped in professional systems and software development organizations?
* Why definitions and actual enactment of particular types of development practices vary among development organizations?
In the second article  the problem that the authors are addressing is that in the existing CTS, amount of time for reacting to a stimuli was much large and it causes emotions is forgotten and won t go through the other learner types. Hence the agent won t be able to be compatible with the environment. That means in current CTS the emotional leaning is not implemented.
The key scientific questions that the author is addressing are:
* How to implement an emotional learning mechanism in CTS in a better way?
* Is there any way to increase the autonomy of a cognitive tutor agent in order to make it similar to human being?
In the third article  the research question addressed by authors is:
* How to achieve better software quality through testing?
The answer is illustrated by proposing 4 main actions to improve testing process:
* A simple first idea
* Requirements engineering
* Implementation of a test center
* Test controlling and performance measurement
In the fourth article  the main question is:
* What technology is the best for people s identities?
It s mentioned on the article that the common technologies that are used are vulnerable to forgery and theft and lapses in user s memory thus face recognition as one of the biometric technologies that have a natural place in smart environment identify people by their physiological characteristics, and in recent years the attention of the scientific community has been focused on face recognition.
1.4 Research Approaches and Methods
In this part, we will discuss about the methods and research approaches used in each of four articles to answer their research questions. Since the first article  is considered as an inductive research and uses a grounded theory approach , the paper does not include any hypothesis; instead it starts based on the conceptual theory and tries to answer why the definitions of the development practices and enacting them vary among organizations; and continues on the basis of observations and case studies. Input data actually is collected from the observations from eight companies. The research methodology used to find the answers of the research question is grounded theory , which best suits because the authors have theoretically answered the questions first and then tried to verify it. For this case the suitable methodology is grounded theory .
In the second article  the goal is to improve the existing system, understanding of the emotional learning and its effects on different parts of CTS. Therefore the research methodology is Qualitative . The method were used in this article is Experiment because it aims to develop the existing system by using hypotheses which is the two-route emotional learning model to improve the CTS. Since the article is an investigation through the other literatures the research approaches is Empirical .
In the third article  the research approach is Empirical and the research method is Action research because it has focused on solving problems and seeks to lead technology-oriented as well as organizational-focused. Qualitative research methods were used. Qualitative research method focus on increasing understanding of a substantive area; it involves methods such as case studies. In this research interviews can be seen as a case study.
In the fourth article  the research approach is Empirical because research is conducted to answer specific question. The research methodology of this article is Qualitative , it concerned with increasing the understanding of how face recognition technology works. The research method is Action research . The article provides general questions and then the authors theoretically answer those questions by gathering all the data and extend experience (add strength) to what is already known through the previous researches. Actually the authors tried to improve the way they address issues and solve problems.
The comparisons of research approaches, methods and methodologies of the four articles summarized in table 1 in following:
Table 1: comparisons of research approaches, methods and methodologies of the four articles
From very early on ISD research has focused on how development should be done (e.g. Dijkstra 1965, Parnas 1972). This resulted in the construction of numerous formal systems development methods (SDMs) (Jayaratna 1994, Avison & Fitzgerald 2002). Adherence to methods was regarded by default as useful and methods thus were to be strictly pre-defined and applied as intended by the method developers (Humphrey & Snyder & Willis 1991, Jarke & Pohl & Rolland & Schmitt 1994). Quite the reverse, a number of ISD researchers have discussed that SDMs experience a great extent of pragmatic adaptation during their adoption and use, if used at all (Stolterman 1992, Fitzgerald 1998, Kautz 2004, Vogelsang & Kensing 2006, P iv rinta & Sein & Peltola 2008). One stream of research even suggested that amethodical ISD exemplifies an alternative view on development practice in modern-day ISD organizations. That is, methods would have no prescribed role at all, while ISD practice emerges through contextual interaction and improvisation (Truex & Baskerville & Travis 2000).
A few Empirical studies have focused on the notion of individual systems developers and how they recognize the relevance of methods and utilize them in their work. These include works on the relationship between developer experience and modes of SDM use, or determinants of developer intentions to use methods (Hardgrave & Davis & Riemenschneider 2003). Education of reflective systems developers with regard to SDMs has also received attention (Mathiassen & Purao 2002). A few studies have discussed ISD project exigencies in relation to local method adoption, where the unit of analysis has been a project (e.g. Kautz 2004, Madsen & Kautz & Vidgen 2006).
A good stream of research has discussed methods in the context of systems development companies and organizations beyond individuals and particular projects. For example, Fitzgerald et al. (2003, 2006) focused on internal software development in large IT industry organizations. Mathiassen & Vogelsang (2005) and P iv rinta et al. (2008) have followed longitudinally how particular development methods have been adopted and adapted in professional development organizations. Beyond the discussion about methods, the reflective systems development (Mathiassen 1998), a.k.a. the professional work practice (Iivari & Lyytinen 1998), approach has focused on improving actual development practices in development organizations. It has focused especially on action research and local improvements based on contextual circumstances (Mathiassen 1998). In the 1980s (Mathiassen 1998), it was one of the first research programs which challenged the belief that development methods as such would improve the effectiveness of ISD (Iivari & Lyytinen 1998).
We shared this focus by asking how and why actual practices are shaped in professional systems and software development organizations. In the early stage, we met the challenge to conceptualize the patterns of defined and enacted ISD practices in our case organizations. Few unifying frameworks for this purpose exist. For example, Fitzgerald, Russo and Stolterman (2002) presented the method-in- action framework, which explain the influence of a number of factors on contemporary ISD practices. It has been used in a number of works, e.g. by Madsen et al. (2006) to describe how a local method emerged in an individual systems development project. Unlike such works as Kautz (2004) and Madsen et al. (2006) which focused only on individual systems development projects, we have an extended view that includes the level of systems development organizations, which, of course, may take part in numerous projects. Software process improvement frameworks such as CMM (Humphrey 1989) build on more detailed models of the development process at the level of the organization. However, their intended use is to assess organizational capabilities by benchmarking the actual ISD practice to a predefined set of key practices, which leaves the practice outside the key practices unexplored.
In the second article  the authors explain about emotional learning models and its effects on different learning. There are other CTS models implemented by others such as Gratch and Marsella(2004), Franklin as well as Vela?squez (Vela?squez, 1996) , proposed their emotional Architectures . But these models have some faults. In the previous models emotions rise and fall quickly and they do not support propagating information and tracking the situation in real time like the system exits in human being.
The third Article  explains that there is no set of best test practices, which is implemented by any company. Some of the recommendations can be found in academic literature. Other practices found in this article are not illustrated for the field of software development.
In the fourth article  ,the authors mention that subject of face recognition is as old as computer vision because of the topic s practical importance and theoretical interest from cognitive scientists. Despite the fact that other identi?cation methods (such as ?ngerprints or iris scans) can be more accurate, face recognition has always been a major research focus because it is noninvasive and seems natural and intuitive to use.
In first article  the report has a discussion part and the conclusion part. The former contains the observations, which are made from the data with the help of NIPO grid. It describes how the scope of real utilization of ISD may differ in different levels of one organization and across organizations. In this section it is also claimed that NIPO framework has some advantages over CMM and other SPI methods. In latter (conclusion) they have mentioned that the construct that they extracted is something that integrates previous SPI approaches into a common framework. Although it is mentioned in abstract that the construct would act as dependent variable for forming the theory it seems that it is not clearly explains how this dependent variable is connected to other variables, while it also seems that the whole text talks about this issue it seems that it would be better to have a clarification of that in conclusion or discussion parts.
In the second article, the authors conclude that by implementing EL the agent will learn and react fast for different kinds of stimuli, which can be used for other learning, and help agent to treats better in a situations whit the existence of different and changeable stimulus.
For the third article conclusion is that this article presented results from a project with regional companies. They built a framework to classify recommendations on how to achieve better quality through testing. They proposed 4 main actions to improve the testing process.
In the fourth article  face recognition technology could play a major part in helping humans as helpful assistants. But to achieve this goal next-generation face recognition system should fit naturally within the patterns of normal human interactions. so future smart environment should use the same models as humans and have the same limitations.
In general, we think that the authors of the article  and have walked almost completely in the framework of a research report. The structure of the reports follows recommendations. Especial thing about the first article  is that it points to the shortcomings of the research in discussion session and further it refers to them as future works in conclusion section. Also article  had some avenues for future research while in the second article there is no hit to that effect.
We would like to thank Bo Helgeson , giving us idea regarding how to write article and other resource information. We are also thankful to Ted Gunnarson and Eva Norling for teaching us how to identify the key words, and how to search scientific text.