ANLY482 AY2016-17 T2 Group 2 Project Overview Methodology
| Background | Data Source | Methodology |
|---|
Tools Used
For data preparation and EDA, the team chose to use JMP software as they are familiar with the usage of this software. To facilitate future extension of the project, the client requested for us to use R programming language for the final outcome. R has a mature and growing ecosystem of open-source tools for mathematics and data analysis.
Methodology
| Data Collection |
Kaiso provided us with transaction records on musical and concert data. These records consist of data from both phone booking and internet booking channels. Apart from these transaction datasets, Kaiso also provided us with customer demographics data and sports matches data. In total, we have obtained 9 datasets from them:
- Lottery transaction data (lottery15.csv, lotteryAug-Oct.csv, lotteryRB.csv)
- Sports transaction data (sports15.csv, sportsAug-Oct.csv, sportsRB.csv)
- Sports matches data (Matches_Master.csv, League name.xlsx)
- Customer demographics data (data_cst.xlsx)
| Literature Review |
To gain more domain knowledge, we will seek to read up on research papers, articles and news related to our area of topic which is ticketing analytics. Furthermore, we aim to focus our reading on online ticketing because we will be using it as our basis when we perform our analysis. In addition, this will provide us with sufficient theoretical knowledge to conduct these analyses.
In this project, we will be conducting comparison analysis on the datasets. Thus, we will also be exploring on papers related to “cross-sectional analysis” and “longitudinal analysis” to aid us in our understanding of this two subjects.
| Data Preparation |
Before performing any further data analysis, the first step is to prepare the data. We will clean the data to handle outliers and missing values. In addition, we will perform data normalization and transformation on the given dataset.
For outliers, we will first determine if the values are due to human or system error. If it is due to human or system error, we can safely remove that transaction from our analysis. Otherwise, we will conduct separate analysis of these outliers values.
For missing values, we will determine the number of missing values. If the number is significant, we will use prediction techniques to predict these values based on the data set. Otherwise, we will remove these transactions from our analysis so that it will not affect our findings.
Lastly, we will perform data normalization and transformation. Some fields in the phone purchasing dataset and internet purchasing dataset have different scales and values even though they represent the same information. Also, due to system changes in Kaiso's IT infrastructure, there are some differences in the way the data is stored and named. Therefore, we will perform data normalization and transformation to ensure that values throughout both dataset are consistent before we can perform any analysis.
| Exploratory Data Analysis (EDA) |
In the initial stage of this project, we will examine the dataset to have a better understanding of the various aspects of the dataset. We will then proceed to perform comparison studies between the datasets. The purpose of the comparison studies is to identify any behavioral differences among the customers. There are two studies which we will be doing - cross-sectional analysis and longitudinal analysis. Some of the comparisons which we will be looking at for both analyses are the frequencies of transactions for account holders in relation to the different ticketing types, the popular time of transaction, type of transaction and amount per transaction.
Cross-sectional Analysis
In this analysis, we will perform comparison study on customers in the same time period of 2015 and 2016. We have 2 months of data for 2016 and will be subsetting the 2015 dataset to contain records from the same time period only.
The purpose of using the same time period for both years is to eliminate any seasonal fluctuations that exists in the datasets.
Longitudinal Analysis
For this analysis, we will be examining the behavioral change of old customers that bought tickers before and after the launch of the online ticketing channel. This analysis aims to answer the question on whether customers purchasing behaviour changed after the launch. Hence, we will filter out data records to include only old customers (customers who registered before the launch).
We will be doing comparisons on data two months before and two months after the launch of the online ticketing site.
| Dashboard |
Following the analysis, an analytical dashboard will be built to visualize our findings. The dashboard will display the key variables of the data and how they affect the customer purchasing behaviour. The customer engaging teams would be able to utilize the dashboard to display and better understand the differences between the customer behaviours before and after the launch of the new system.
The dashboard will use a framework that allows Kaiso to update their dashboard by uploading their dataset every time they have a new dataset. Design, statistics and visualization will be our main considerations when building the dashboard so that they can easily unveil the differences that they are looking for.
| Recommendations & Insights |
From our analysis and dashboard, we seek to assist Kaiso in understanding the characteristics of their customers. We will be proposing business strategies and recommendations to them based on the insights that we have uncovered.
