Difference between revisions of "ANLY482 AY2016-17 T2 Group09: Project Overview"
Tjwang.2013 (talk | contribs) |
|||
| Line 57: | Line 57: | ||
<font size =3 face=Georgia > | <font size =3 face=Georgia > | ||
| − | LKS library is the main study area for SMU students. | + | LKS library is the main study area for SMU students. The management team of Li Ka Shing Library is striving for better user experience. Unlike e-book usage or search request which data can be easily collected, they have little information about physical usage of the library, especially the usage of specific user groups. Our group has been studying in SMU for four years,and we are motivated to know how our peers are using the library as well. |
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
</font> | </font> | ||
| Line 83: | Line 75: | ||
Compared with EDA data visualisation, statistical test in terms of one-way ANOVA brings a number of advantages and great statistical power due to increased precision and more informative interpretation of the results. It provides us with deep insights on the relationships between different variables and statistically prove our hypothesis based on the data. In our case, we use one-way ANOVA test to further validate the conclusion we drawn from EDA process. With our statistical test result, we reject or accept our null hypothesis, and this provides us with statistical support for our final conclusion. | Compared with EDA data visualisation, statistical test in terms of one-way ANOVA brings a number of advantages and great statistical power due to increased precision and more informative interpretation of the results. It provides us with deep insights on the relationships between different variables and statistically prove our hypothesis based on the data. In our case, we use one-way ANOVA test to further validate the conclusion we drawn from EDA process. With our statistical test result, we reject or accept our null hypothesis, and this provides us with statistical support for our final conclusion. | ||
| − | |||
| − | |||
</font> | </font> | ||
</div> | </div> | ||
Revision as of 00:08, 24 April 2017
Sponsor Background
Li Ka Shing Library (LKSLIB) is the first library of Singapore Management University, officially opened on 24 February 2006. The Library is named after Hong Kong businessman Dr. Li Ka-shing, and the Li Ka Shing Foundation donated and endowment to the library for collections. The main objective of the library is to offer an interactive study and research space for SMU community.
The LKSLIB includes four floors that comprise about 8,800 square meters with 1,800 seats. Inside the library, there are a variety of spaces including open spaces for individual and collaborative use, learning commons which opens 24/7, quiet areas that for individuals to focus on their work, project rooms with LCD panels, investment studio, postgraduate lounges etc. As a modern library, it is also well equipped with high-speed wireless network, color printers, scanners, public computers with professional financial software available, up-to-date newspapers and magazines, collections of lifestyle videos and games, and this is also the reason why LKSLIB is so attractive for SMU community.
Project Background
In our project, our focus is on analyzing the library entry information from the card reader logs. The card readers are located at the entrance of the library gantries, both located at the main entrance of LKSLIB and at the linkbridge side entrance. Students need to tap their card whenever they enter the library. This provides us with the entry information, which includes timestamp and basic information about the student. To better understand the library usage, the library management team is interested to know whether we could find any usage pattern for library of a particular user group (e.g. Dean’s List student), and see if any other business insights could be drawn from the data. We will also work on statistical analysis in order to confirm on our insights.
We use R to build a web application to clean the raw data and use Tableau to do data visualization to compare the usage level for dean’s list and non-dean’s list students, Singaporean students and international students. Then, we will do one-way ANOVA confirmative analysis using SAS JMP.
Motivations
LKS library is the main study area for SMU students. The management team of Li Ka Shing Library is striving for better user experience. Unlike e-book usage or search request which data can be easily collected, they have little information about physical usage of the library, especially the usage of specific user groups. Our group has been studying in SMU for four years,and we are motivated to know how our peers are using the library as well.
Objectives
We aim to study the correlation between the two sets of variable:
- Dean's Lister vs. Non-Dean's Lister
- Singaporean Students vs. International Students
Exploratory data analysis (EDA) is an approach to analysis data sets to summarise their main characteristics. Bar charts are usually used to present the grouped data with rectangular bars with lengths proportional to the values that they represent in order to show the comparison among categories. Primarily EDA is often used for seeing what the data can tell us and explore the data to see the possible hypotheses that could lead to model fitting and hypothesis testing. In our study, we use EDA to have an overview of the analysis and draw a primary conclusion from that.
Compared with EDA data visualisation, statistical test in terms of one-way ANOVA brings a number of advantages and great statistical power due to increased precision and more informative interpretation of the results. It provides us with deep insights on the relationships between different variables and statistically prove our hypothesis based on the data. In our case, we use one-way ANOVA test to further validate the conclusion we drawn from EDA process. With our statistical test result, we reject or accept our null hypothesis, and this provides us with statistical support for our final conclusion.
