Decision Support Systems Project | Excel, Minitab, & Cognos | Review Attached PDF
ponch75
MIS 648
Group Project
Abdulrahman Alrowais, Narendranath Singh, Suhas Sonawane
Abstract:
The project comprises of the implementation of Decision Support Systems principles. Business
Intelligence is the key factor in DSS. BI helps making better decisions by applying the modeling
techniques using BI tools. There are numerous tools available for BI. This project uses MS
Excel, Minitab, MS Access and Auspicate as tools. The project is based upon the idea of helping
students of NJIT to register for the particular course under certain professor which they can find
rating as helpfulness, easiness, clarity of that respective professor. In BI terms the project is
mainly dealing with gathering information, putting it together to have a meaningful model and
project useful information as output which eventually would help make best decisions. Project
deals with data from NJIT consolidated with data from ratemyprofessors.com. Together it gives
tabular model and applying BI techniques on the model.
Part I
Definition of Problem:
Every year during the start of the semester students are confused about the courses they should
register and under which professor. According to Miranda and Saunders (2003), people who
engage in online dealings are much more likely to be more responsive and considerate on the
quality of information a website presents since information quality provide them a useful
approach in coming up a decision whether they will go with or not. Student find it difficult to
analyze it under one system, generally has to go to senior or look into website for grading and
ranking of professors. Lecture timing and the course selection get difficult to adjust for the
students. We try to build an intelligent system, where students can review ranking, courses and
timing before registering, using which one can easily analyze the course they want to register.
Useful and high quality contents in a BI system must be provided in order to attract practical
users to utilize it (Tung et al. 2009). Using the same system the school management can analyze
the progress of the faculty and the student results. The results of students and ratings of the
management school professors has to be accurate and from the source, management school will
provide the student results and on the other hand students will rate their professors. For Wang
(2008), reliability is the ability of the system to provide accurate and dependable service; thus
establishing the importance of the quality of information integrated in the system in order to
develop reliability.
Justification:
The system takes help of Minitab tool which helps in Data Mining and Business Intelligence.
Since student wants to take best decision for his management course, he can take the help of our
system. In the system student can check for the best classes under the best professor. Student also
can look up for the time to attend the classes. All these opportunities typically impacts not just
our BI system itself but also student’s decision-making patterns. Moreover, quality of
information in a DSS or a decision support system allows the decision maker to justify the
decision choices, arguing that if the used information is timely, accurate and reliable, then any
decision made is effective (Medina and Chaparro 2008).
Benefits:
a. It categorizes the information from the data in various different comparisons.
b. Using Minitab, different kinds of operations can be performed.
c. Student can generate charts for the best rated professors.
d. Report can be generated by considering course number, sections and time.
e. Grades of the professors can be viewed in charts, tables and different types of statistics to
understand better about the course.
f. Student can perform cross tabulation and regression to get the most desired information.
Course Integration:
The concept actually integrates with the topics such as Business intelligence and cognitive
decision making. The data is being generated from NJIT student services. Then this data is
further being integrated with rating of the professor and grades which is generated from the 2
different websites. This integration helps us to generate desired statistical data in Minitab.
The data that we are going to use in our intelligence system is a categorical data about
professors and their management classes. The data will contain information (course number,
section, days, time, etc.) about each professor and each one of their classes for the fall semester
2012. We will get the data from three different sources and they are:
NJIT Student services: course information (course number, section, days, time, etc.)
ratemyprofessors.com: professor’s ratings and comments by their students.
myedu.com: course grades for classes and average grades for professors.
Here is a descriptive statistics of the data we have so far and a sample graphical summery of one
of the variables:
Variable N N* Mean SE Mean StDev Minimum Q1 Median Q3
CRN 48 2 23432 48.7 337 23298 23309 23322 23334
Crse 48 2 446.6 24.6 170.5 190.0 296.5 480.0 620.0
Sec 36 14 148.7 36.3 217.5 1.0 1.0 101.0 101.0
Cred 48 2 2.9583 0.0417 0.2887 1.0000 3.0000 3.0000 3.0000
Hours 46 4 1.821 0.210 1.425 0.000 0.938 1.250 3.500
Cap 48 2 34.69 3.34 23.16 0.00 30.00 30.00 50.00
Act 48 2 30.31 3.24 22.43 0.00 13.00 29.00 40.00
Rem 48 2 4.38 1.54 10.67 -23.00 0.00 2.00 10.75
Variable Maximum
CRN 24569
Crse 699.0
Sec 851.0
Cred 3.0000
Hours 3.500
Cap 70.00
Act 74.00
Rem 32.00
700600500400300200
Median
Mean
500480460440420400
1st Q uartile 296.50
Median 480.00
3rd Q uartile 620.00
Maximum 699.00
397.11 496.10
390.00 492.00
141.90 213.52
A -Squared 1.14
P-V alue < 0.005
Mean 446.60
StDev 170.46
V ariance 29057.56
Skewness -0.04411
Kurtosis -1.20661
N 48
Minimum 190.00
A nderson-Darling Normality Test
95% C onfidence Interv al for Mean
95% C onfidence Interv al for Median
95% C onfidence Interv al for StDev
95% Confidence Intervals
Summary for Crse
Part II
Every year during the start of the semester students are confused about the courses they should
register and under which professor. Student find it difficult to analyze it under one system,
generally has to go to senior or look into website for grading and ranking of professors. Lecture
timing and the course selection get difficult to adjust for the students. We try to build an
intelligent system, where students can review ranking, courses and timing before registering,
using which one can easily analyze the course they want to register. Using the same system the
school management can analyze the progress of the faculty and the student results.
1. Build an Access or Excel model. Build Mini Tab and Auspicate data model.
2. Write report on the above model.
Summary: The dataset we used is to analyze the different queries to understand the NJIT
course structure model. We used various websites to get our data, www.ratemyprofessor.com
, www.njit.edu .
Research Questions: The model is to derive few queries which was analyzed using Access
and Auspicate system.
We have divided possible queries which would provide solution to students/
professors also Management could decide how they could improve the course
structure for future. Few queries as follows:
1. BI use for Students:
Analyze which attribute of professor is the driving variable for student to select a
particular course.
Solution: We analyzed this query in Minitab and Auspicate , first we compared the easiness
of the professor to the clarity of the professor for all the courses offered by NJIT. We see the
changes in the graph output where the student selects the easiness and also the clarity as
different criteria for different courses. Later we tried comparing overall performance with the
easiness of the course, we had our finding that the overall performances of professors are
most driven by the easiness of the course. The student generally prefer the ease the course is
and it the driving criteria for selecting a course.
Instructor
H el p f
ul ne
ss
Ea sin
es s
302928272625242322212019181716151413121110987654321302928272625242322212019181716151413121110987654321
5
4
3
2
1
0
D a
ta
Chart of Easiness, Helpfulness
Percent within all data.
Instructor
O v e
ra ll q ua
lit y
Ea sin
es s
302928272625242322212019181716151413121110987654321302928272625242322212019181716151413121110987654321
5
4
3
2
1
0
D a
ta Chart of Easiness, Overall quality
Percent within all data.
Analyze the best overall performance of the professor using least number of hours to teach a
course.
To analyze we created query in Access as well as Auspicate to compare our finding. The
realization of our find was, students generally prefer online courses over classroom also they
prefer classes on Tuesdays. The result in Access shows us the popularity of the course as
well. There was more number of registrations for a particular class which is show in the data
result.
There use of the BI tool for Administrator:
The administration team at School of Management wants to analyze which day or time would be
best to offer a course. Here the administration would analyze each course at a time. For our
analysis we picked a single course “Principle of Management.”
After looking at the result , the Management want to decide who would be the best professor to
teach a course, so as the student could get best results. The management wants to understand
which professor would be best for the selected course.
The result shows the Professor ‘9’ works best on day ‘2’ with overall performance ranking of 5.
So Management deciding to keep the Professor for that particular course for future.
Contribution:
Business Intelligence basically helps us to make proper decisions. We used the Business
Intelligence tools such as MS Excel, MS Access, Minitab and Auspicate. We gathered the data in
Excel. Data mining and BI has helped companies to spot patterns, bring product offering to right
customers and nurture customer relationship. As the companies expand their web customers, the
use of BI is further more mined for customer relationship. BI helps consolidate, analyze, and
provide access to vast amount of data for business analysis and decision making.( Mazon, J.
Garrigos, I. Daniel, F. and Trujillo, J. 2012). We modified it according to the instructions given
and processed the data in Access to perform various type of queries. Access helped in organizing
the data and getting outputs providing useful information. Minitab is higher level tool which lets
us apply regression which is a type of data mining method. We also got to output various graphs
in most accurate way. Analysis of variance tool in Minitab helped us understand the variation in
driving dependant variables. This whole process of Business Intelligence made us understand the
data and taught us how to get useful information out of it which actually makes sense. Auspicate
is the Browser based tool which also helped us performing BI operations on cloud. It teaches us
how the data is stored and processed on cloud. It has one of the fines GUIs and the functions are
user friendly.
Summary:
The project of development of BI systems for students and administration to help them analyze
the scenarios depending on their constraints, make better decisions that would help them figuring
out what choices they should make. The students can select the best professor according to their
constraints. Administration can help serve students (Users) to use the systems and get better
output. The tools used in the process are MS Excel, MS Access, Minitab and Auspicate.
However we struggled a bit in making better use of Auspicate o generate critical reports and
models. Minitab and MS Access hand in hand helped fulfilling the requirements which helped us
reaching the project goals.
References:
Miranda, S.M. and Saunders, C.S.,(2003). The social construction of meaning: an
alternative perspective on information sharing. Information Systems Research, 14
(1), pp. 87–107.
Tung, F.C., et al. (2009). An extension of financial cost and TAM model with
IDT for Exploring Users’ behavioral intentions to use the CRM information
system. Social Behavior and Personality, 37(5), pp. 621-626
Wang, M.Y. (2008). Measuring e-CRM service quality in the library context: a
preliminary study. The Electronic Library, 26(6), pp. 896-911
Medina, M.Q. and Chaparro, J.P. (2008). The impact of the human element in the
information systems quality for decision making and user satisfaction. The
Journal of Computer Information Systems. 48(2). pp.44-52
Mazon, J. Garrigos, I. Daniel, F. and Trujillo, J. 2012, “Introduction to the special
issue of Business Intelligence and the Web,” Decision Support Systems (52:4),
March, pp. 851-852
"RateMyProfessors.com." New Jersey Institute of Technology. N.p., n.d. Web. 11
Dec. 2012.
"Business Intelligence & Modeling Systems." Synergy. N.p., n.d. Web. 11 Dec.
2012.
"Demo: Exploring a Sample Dimensional Data Model." Demo: Exploring a
Sample Dimensional Data Model. N.p., n.d. Web. 11 Dec. 2012.
"Welcome to Business Modelling Associates." Business Analytics, Predictive
Modelling, Supply Chain Strategy. N.p., n.d. Web. 11 Dec. 2012.
"Power, Politics, and MIS Implementation." Power, Politics, and MIS
Implementation. N.p., n.d. Web. 17 Dec. 2012.
"Knowledge Management and Data Mining for Marketing." Knowledge
Management and Data Mining for Marketing. N.p., n.d. Web. 17 Dec. 2012.
Mintzberg, Henry, Duru Raisinghani, and André Théorêt. JSTOR. N.p.,
n.d. Web. 17 Dec. 2012.