Business Report for Predict & Prescript Analytics
HOTEL Chrysalis
1
,Executive Summary
Mission statement
The Las Vegas Hotel wishes to expand its operations. As a result, an analysis is required to
determine the most significant character of the hotel so that the company can determine
the method of effective hotel renovation. Another mission is the optimal optimisation,
arrangement of renovating the casino and business strategic restrictions. The hotel's
primary focus is to ensure that guest reach is maximised while also increasing customer
satisfaction.
Analysis Methodology
This advertising project was planned to use the CRISP-DM (Cross-industry Standard Process
for Data Mining) methodology. The historic dataset was analysed while it was being
transformed and cleaned. Furthermore, descriptive analysis was used to comprehend
several post characteristics in five dimensions: content type, time, post-performance,
consumer and consumption, and interaction performance. During the predictive analysis
phase, a multiple linear regression model and cross-validation method were designed.
Furthermore, the optimised result was carried out by linear programming with Excel
software and Rstudios during the prescriptive analysis phase.
Recommendation from the results
Recommendations are made based on the findings of various analyses, including database
update synchronisation, guest optimisation strategy, and the potential effects of external
factors such as swimming pool development and Gym. Several analysis approaches, such as
regression model, decision theory of hotel facilities, and promoting the delivery of high-
quality services, are also suggested.
2
, 2. Introduction
The primary goal of this report is to determine the best strategies for renovating the hotel's
various locations. This report performed descriptive, predictive, and prescriptive analysis
based on the data sources gathered. Furthermore, data transformation, multiple linear
regression, linear optimization, and cross-validation are among the analysis measures used
in this report. Furthermore, the structure of this report is primarily as outlined below. This
report also includes an external R script file and an excel file containing a sensitivity report
and descriptive statistics.
• Data organisation and descriptive analysis
• Analysis and Conclusion
• Recommendation
• Predictive analysis: design of multiple linear regression models
• Prescriptive analysis: linear optimization
The following issues are addressed in this report:
The optimization of a renovation plan based on the opinions of multiple reviewers
on existing facilities.
Predictive trends of renovation arrangement from a dataset of historical customer
activity
The major influencing factors of visitor activities
The maximum number of returning visitors solution.
3. Data Structure and Descriptive Analysis
3.1. Data Structure
This structured dataset gathered from the hotel is made up of twenty attributes and five
hundred and four records (Refer Table 1). This dataset's data types are primarily ratio and
nominal. The dataset's average completeness rate is around 95%. Furthermore, no primary
key is used to represent each record.
3.2. Data Cleaning and Transformation
For starters, this spreadsheet has a significant incompleteness problem. Figure 1
shows several blank cells in the "User continent," "Member years," "Review month,"
and "Review weekday" columns. To avoid analysis bias, it is suggested that these five
records with no value be determined. However, because these columns may have no
effect on the analysis, they should not be cleaned in the dataset. At this point, 500
records have been selected for further analysis. Second, it is shown that there is a
cell in members years that shows 1800 years that needs to be corrected. Thirdly, the
3
HOTEL Chrysalis
1
,Executive Summary
Mission statement
The Las Vegas Hotel wishes to expand its operations. As a result, an analysis is required to
determine the most significant character of the hotel so that the company can determine
the method of effective hotel renovation. Another mission is the optimal optimisation,
arrangement of renovating the casino and business strategic restrictions. The hotel's
primary focus is to ensure that guest reach is maximised while also increasing customer
satisfaction.
Analysis Methodology
This advertising project was planned to use the CRISP-DM (Cross-industry Standard Process
for Data Mining) methodology. The historic dataset was analysed while it was being
transformed and cleaned. Furthermore, descriptive analysis was used to comprehend
several post characteristics in five dimensions: content type, time, post-performance,
consumer and consumption, and interaction performance. During the predictive analysis
phase, a multiple linear regression model and cross-validation method were designed.
Furthermore, the optimised result was carried out by linear programming with Excel
software and Rstudios during the prescriptive analysis phase.
Recommendation from the results
Recommendations are made based on the findings of various analyses, including database
update synchronisation, guest optimisation strategy, and the potential effects of external
factors such as swimming pool development and Gym. Several analysis approaches, such as
regression model, decision theory of hotel facilities, and promoting the delivery of high-
quality services, are also suggested.
2
, 2. Introduction
The primary goal of this report is to determine the best strategies for renovating the hotel's
various locations. This report performed descriptive, predictive, and prescriptive analysis
based on the data sources gathered. Furthermore, data transformation, multiple linear
regression, linear optimization, and cross-validation are among the analysis measures used
in this report. Furthermore, the structure of this report is primarily as outlined below. This
report also includes an external R script file and an excel file containing a sensitivity report
and descriptive statistics.
• Data organisation and descriptive analysis
• Analysis and Conclusion
• Recommendation
• Predictive analysis: design of multiple linear regression models
• Prescriptive analysis: linear optimization
The following issues are addressed in this report:
The optimization of a renovation plan based on the opinions of multiple reviewers
on existing facilities.
Predictive trends of renovation arrangement from a dataset of historical customer
activity
The major influencing factors of visitor activities
The maximum number of returning visitors solution.
3. Data Structure and Descriptive Analysis
3.1. Data Structure
This structured dataset gathered from the hotel is made up of twenty attributes and five
hundred and four records (Refer Table 1). This dataset's data types are primarily ratio and
nominal. The dataset's average completeness rate is around 95%. Furthermore, no primary
key is used to represent each record.
3.2. Data Cleaning and Transformation
For starters, this spreadsheet has a significant incompleteness problem. Figure 1
shows several blank cells in the "User continent," "Member years," "Review month,"
and "Review weekday" columns. To avoid analysis bias, it is suggested that these five
records with no value be determined. However, because these columns may have no
effect on the analysis, they should not be cleaned in the dataset. At this point, 500
records have been selected for further analysis. Second, it is shown that there is a
cell in members years that shows 1800 years that needs to be corrected. Thirdly, the
3
