cs504049 business intelligence systems

The method of making this report is that each student needs tofirmly grasp and understand the parts learned to apply to the midterm report, such as:Definition of Business Intelligence, M

CS504049 - BUSINESS INTELLIGENCE SYSTEMS

FINAL PROJECT

Instructor: THẦY DƯƠNG HỮU PHÚCExecutor: VÕ QUỐC KỲ – 520H0082

HOÀNG NGỌC NGỌ – 520H0388NGUYỄN NHẤT DUY – 520H0219Course : 24

HO CHI MINH CITY, 2023

CS504049 - BUSINESS INTELLIGENCE SYSTEMS

FINAL PROJECT

Instructor: THẦY DƯƠNG HỮU PHÚC Executor: VÕ QUỐC KỲ – 520H0082HOÀNG NGỌC NGỌ – 520H0388NGUYỄN NHẤT DUY – 520H0219Course : 24

HO CHI MINH CITY, 2023

We would like to thank Ton Duc Thang University, the teachers in the Facultyof Information Technology for their dedication to imparting and supplementing us withmany new sets of knowledge to make the report better In the process of makingassignments and reports will inevitably make mistakes, we hope to receive suggestionsfrom teachers.

We would like to sincerely thank Mr Duong Huu Phuc, who is currentlyteaching Business Intelligence Systems for the group of 04 courses K24 a lot becausehe taught us very passionately, very enthusiastic and easy to understand to be able tocomplete this report well

PROJECT COMPLETEDAT TON DUC THANG UNIVERSITY

We hereby declare that this is my own project and is guided by Mr Duong HuuPhuc The research contents and results in this topic are honest and have not beenpublished in any form before The data in the tables for analysis, comments andevaluation are collected by the author himself from different sources, clearly stated inthe reference section.

In addition, the project also uses a number of comments, assessments as well asdata of other authors, other agencies and organizations, with citations and sourceannotations.

If we find any fraud, we will take full responsibility for the content of ourproject Ton Duc Thang University is not related to copyright and copyright violationscaused by us during the implementation process (if any).

Ho Chi Minh City, December 5, 2023 Author

(Sign and write your full name)

Võ Quốc KỳHoàng Ngọc NgọNguyễn Nhất Duy

TEACHER'S CONFIRMATION AND ASSESSMENT SECTION

Instructor endorsement

Ho Chi Minh City, , 2023 (Sign and write your name)

Evaluation section of the teacher marking the test

Ho Chi Minh City, , 2023 (Sign and write your name)

Our research problem for this report is on Business Intelligence Systems Thepurpose of this report is to address the midterm exam and its reporting is required bythe IT department The method of making this report is that each student needs tofirmly grasp and understand the parts learned to apply to the midterm report, such as:Definition of Business Intelligence, Modeling in Business Intelligence, DataProvisioning, Decision Tree Learning (Classification, Regression), Bayesian Learning,Data Mining for Cross-Sectional Data, The results of the exercise must be absolutelycertain and accurate, solving all 7 part given by the faculty The BI subject report isalso a test that helps the faculty see the learning ability of students in the first semesterof this subject.

1.1 Introduction my group member 10

1.2 Final Report Introduction: 11

1.3 House Rentel Website introduction: 11

2 CHAPTER 2: DETERMINE THE TOPIC: 11

2.1 Topic overview 11

2.1.1 Key Features of House Rental Websites: 11

2.2 Purpose of choosing this topic 13

2.3 Meaning of the topic 13

2.4 Goals and future directions 14

3 CHAPTER 3: DATA COLLECTION AND INTEGRATION: 15

4 CHAPTER 4: DATA CLEANSING AND TRANSFORMATION: 16

5 CHAPTER 5: EXPLORATORY DATA ANALYSIS: 16

6 CHAPTER 6: DATA MODELING AND ANALYSIS: 17

6.1 Describe the problem 17

6.2 What is Linear Regression? 17

6.3 Prediction steps to build a house price prediction model using Linear Regression 17

6.4 Build a prediction model using the Scikit – Learn library 18

7 CHAPTER 7: INTERACTIVE DASHBOARD CREATION: 20

8 CHAPTER 8: DECISION SUPPORT AND RECOMMENDATIONS: 28

8.1 Code for web: 28

8.3 Results: 30

REFERENCES 32

LIST OF ABBREVIATIONS

ABBREVIATIONS

BI Business Intelligence SystemsEDA Exploratory data analysis

LIST OF FIGURES

Figure 1 Heat map 16

Figure 2 DF_NAME 18

Figure 3 Code 18

Figure 4 Code for Linear Regression 18

Figure 5 Code for fit method 19

Figure 6 Result 20

Figure 7 Month, Date (May - July) 21

Figure 8 Find the city with the highest total number of purchases 22

Figure 9 Figure 10 Find area vs price (Floors and Bedrooms) 22

Figure 11 Find area vs price (Bathrooms) 23

Figure 12 Heat map 24

Figure 13 Predict trends for market prices in increasing or decreasing months 24

Figure 14 Check view 25

Figure 15 Area vs pricef 26

Figure 16 Heat map 27

Figure 17 City with the highest total revenue (the darker the color, the higher) 27

Figure 18 Code for web 29

Figure 19 Code for web 30

Figure 20 HOME 30

Figure 21 Results 31

LIST OF TABLES

Table 1 My group member 10

ASSIGNMENT

1 CHAPTER 1: INTRODUCTION1.1 Introduction my group member

01 Võ Quốc Kỳ

Wordpress,

Word Create and edit Microsoft Word Assist in the team on general issues Perform the EDA part Build Website via Wordpress.

Ngọc Ngọ Tableau,Machinelearning

Fulfill the requirements of the assignment, Tableau, support teammates in the team.

03 Nguyễn Nhất Duy

Tableau Mainly grasp Tableau issues and provide general support within the team Always do your role well and help with difficult questions from the lecturer.

Table 1 My group member1.2 Final Report Introduction:

The final project for CS504049 offers students a practical avenue toapply the acquired concepts, tools, and techniques within the course to addressreal-world business challenges Participants are tasked with selecting a datasetfrom diverse sources, engaging in data cleaning and preparation processes, andsubsequently employing a business intelligence tool to craft visualizations,dashboards, and reports, supporting decision-making endeavors The project'soverarching objective is to afford students an opportunity to demonstrate theirproficiency in data collection, analysis, and visualization, thereby generatingmeaningful insights essential for strategic decision-making processes in abusiness context.

1.3 House Rentel Website introduction:

A House Rental Website serves as a digital marketplace connecting property owners with tenants It features property listings with details like location, amenities, and rental terms Users can create accounts, use search filters, and communicate through a messaging system Reviews and ratings contribute to community trust The platform often integrates maps, online applications, and secure payment systems It streamlines the rental process by providing legal documents and notifications The business model may involve fees for property listings or a percentage of rental transactions, creating a transparent and efficient platform for house rentals.

2 CHAPTER 2: DETERMINE THE TOPIC:2.1 Topic overview

2.1.1 Key Features of House Rental Websites:- User Registration and Profiles- Search and Filters

- Property Listings

- Map Integration- Communication Tools- Application and Screening- Reviews and Ratings

2.1.2 Benefits of House Rental Websites:

- Accessibility: Renters gain access to a wide range of rentalproperties, reducing the need for traditional methods like classifiedads or physical property visits.

- Efficiency: The online platform accelerates the rental process,allowing users to browse, inquire, and apply for properties fromanywhere with an internet connection.

- Wider Audience: Landlords can reach a broader audience,increasing the visibility of their rental properties and reducingvacancy periods.

- Time and Cost Savings: Both tenants and landlords save time andresources by leveraging the online platform for property searches,applications, and lease management.

2.1.3 Challenges and Considerations:

- Scams and Fraud: House rental websites need robust verificationprocesses to prevent scams and ensure the legitimacy of listings.- Data Security: Protecting sensitive information, such as personal

details and financial data, is crucial to maintaining user trust.- Regulatory Compliance: Adhering to local housing regulations and

laws is essential for the success and legality of house rentalwebsites.

- Property Management Integration: Integrating propertymanagement tools and services can enhance the overall rentalexperience for both landlords and tenants.

2.2 Purpose of choosing this topic

- Industry Insight: Explore dynamics in real estate and propertymanagement.

- Market Trends: Understand current trends in the real estate andrental sectors.

- Technology Impact: Examine how technology has transformedthe traditional rental process.

- User Experience: Assess how these platforms enhance theexperience for tenants and landlords.

- Entrepreneurship: Identify business opportunities and challengesin the proptech space.

- Consumer Perspective: Understand how house rental websitessimplify property transactions.

- Academic or Research Focus: Investigate digital platforms, commerce, and technology's impact on industries.

e Policy and Regulation: Examine how these platforms navigatelegal considerations and housing laws.

- Innovation and Disruption: Explore how house rental websitesdisrupt traditional real estate practices.

- Global and Local Perspectives: Consider both global trends andlocal nuances in the house rental market.

2.3 Meaning of the topic

Exploration and understanding of online platforms that facilitate therental of residential properties It involves examining the features,

benefits, challenges, and overall impact of websites that connectlandlords offering rental homes with tenants seeking accommodation.The topic delves into the technological, economic, and societal aspects ofhow these platforms have transformed and streamlined the process offinding, listing, and leasing houses Whether approached from anindustry, market, technological, entrepreneurial, consumer, academic,regulatory, or innovation perspective, the topic aims to provide insightsinto the dynamics of the evolving real estate and rental landscape in thecontext of digital platforms.

2.4 Goals and future directions- Innovation Focus:

o Goal: Drive ongoing proptech innovation.

o Future: Explore VR, AI, and blockchain applications.- User-Centric Approach:

o Goal: Prioritize user satisfaction.

o Future: Implement features based on user feedback.- Market Expansion:

o Goal: Increase user base diversity.

o Future: Expand to new locations with tailored offerings.- Regulatory Compliance:

o Goal: Ensure adherence to local regulations.o Future: Stay updated, collaborate with legal experts.- Sustainability:

o Goal: Promote eco-friendly practices.

o Future: Encourage energy-efficient property listings.- Data Security:

o Goal: Strengthen data security measures.o Future: Invest in cybersecurity technologies.

- Community Building:

o Goal: Foster user community.

o Future: Introduce forums and events for connections.- Property Management Integration:

o Goal: Facilitate seamless management.

o Future: Collaborate with property management providers.- Smart Home Integration:

o Goal: Embrace smart home technologies.o Future: Partner with IoT and automation providers.- Research and Development:

o Goal: Stay ahead in industry trends.

o Future: Invest in R&D for continuous improvement.3 CHAPTER 3: DATA COLLECTION AND INTEGRATION:

Students will identify relevant data sources and employ techniques to gatherdata from various internal and external sources They will then integrate the collecteddata into a single source such as centralized repository or data warehouse for furtheranalysis.

- Delete unnecessary columns

- Check to see if any values are null

- Thereby, we can analyze how the columns' values correlate witheach other Basically, the price column has the most correlation

with the remaining columns, proving that those factors have moreor less impact on house prices.

Figure 1 Heat map

4 CHAPTER 4: DATA CLEANSING AND TRANSFORMATION:In this phase, students will apply data cleansing techniques to ensure dataquality and remove any inconsistencies, errors, or outliers They will also transform thedata into a suitable format for analysis, ensuring data compatibility and uniformity.

5 CHAPTER 5: EXPLORATORY DATA ANALYSIS:

Students will perform exploratory data analysis techniques to uncover patterns,trends, and relationships within the dataset They will use descriptive statistics, datavisualization, and other analytical methods to gain insights and identify key businessmetrics.

6 CHAPTER 6: DATA MODELING AND ANALYSIS:

Based on the exploratory analysis, students will design and implementappropriate data models to facilitate advanced analysis They will utilize statisticaltechniques, predictive modeling, and data mining algorithms to extract meaningfulinformation from the data.

Building a model to support house price prediction using Linear Regression 6.1 Describe the problem

Set in America, we will act as an agent country to predict houseprices for regions With the dataset already prepared, The current taskis to use a linear regression model to be able to calculate How muchwill the house sell for?

6.2 What is Linear Regression?

- Linear regression is a statistical method for regressing data with thedependent variable having continuous values while the

independent variables can have either continuous values or categorical values Linear regression is one of two major forms of supervised learning based on sample data sets.

- In other words "Linear Regression" is a method to predict the dependent variable (Y) based on the value of the independent variable (X) It can be used for cases where we want to predict a continuous quantity For example, predicting traffic in a retail store, predicting how long users spend on a certain page or the number of pages visited on a certain website, etc

6.3 Prediction steps to build a house price prediction model using Linear Regression

- Find data: Here we use the dataset on the platform: usa_house.csv- Data processing: Clean data, handle missing values,

- Data separation: for training and testing to build and evaluate the model.

- Model building: Create a linear regression model to understand the relationship between characteristics and house prices.- Perform model evaluation: Evaluate the model's performance on

the test set using metrics such as MSE or RMSE.

- Deployment and Prediction: Deploy powerful models into world applications to predict house prices based on user input.6.4 Build a prediction model using the Scikit – Learn library

real-6.4.1 Split data into train and test

- Now let's start training the regression model First, we will need to split our data into an array We will remove the Address column because it only has text information that the linear regression model cannot use

Figure 2 DF_NAME

- Now we have two variables x and y as required by the model These two variables are based on the data that is the dataset we have to train the model Now we separate the above variables into train and test values, these two values we will always encounter and use in the process of building machine learning models First, from the Scikti – Learn model_selection library, we import train_test_split, this method helps us create a regression model

Figure 3 Code

- Then we create 4 variables, including X_train, y_train and X_test, y_test With the input argument being the X and y values we took from the data above, test_size returns the percentage of data divided, for example 0.2 corresponds to data divided into 20% of the value test, The rest is train data random_state is equal to a corresponding number to ensure that every time we run the model again, the random split value received is the same You can give it any number.

6.4.2 Create a model and train Linear Regression

- From the Scikit – Learn library, linear_model imports module LinearRegression