Factors affecting purchase decisIons by customer ứr0ups
Suitable Apartment Size: One of the primary considerations for customers when purchasing real estate is the size of the apartment Different customer groups have varying requirements based on their family size, lifestyle, and future plans For example, young professionals or small families may prefer smaller apartments that are more affordable and easier to maintain On the other hand, larger families may seek spacious apartments with multiple bedrooms to accommodate their needs
Surrounding Infrastructure: The availability of surrounding infrastructure plays a crucial role in purchase decisions Customers consider factors such as proximity to recreational facilities, waste management facilities, green spaces, and public amenities Living near parks, playgrounds, and other recreational areas is particularly desirable for families with children Access to quality schools, workplaces, markets, supermarkets, and shopping centers also influences purchase decisions.
Proximity to Schools and Workplaces: The location of a property in relation to schools and workplaces is an important consideration for customers Families with school-going children prefer properties that are close to reputable educational institutions Similarly, professionals seek properties that offer convenient access to their workplaces, reducing commuting time and enhancing work-life balance
Accessibility to Markets and Commercial Centers: The proximity to markets, supermarkets, and commercial centers is another crucial factor influencing purchase decisions Customers prefer properties located near vibrant markets or commercial hubs where they can easily access essential goods and services This convenience adds value to the property and enhances the overall living experience
Community Environment: The community environment surrounding a property significantly impacts purchase decisions Customers consider factors such as the quality of neighbors, safety, and overall community atmosphere A well-maintained and secure neighborhood with a strong sense of community can greatly influence the decision to purchase a property
Proximity to Main Roads and Highways: Easy access to main roads and highways is an essential factor for many customers Properties located near major transportation routes provide convenience in terms of commuting and accessibility to other parts of the city or region This factor is particularly important for individuals who rely on public transportation or frequently travel by car
Attractive Interior Design: The interior design of a property is an influential factor in purchase decisions, especially for those seeking aesthetic appeal Customers often prefer well-designed apartments with modern finishes, functional layouts, and high-quality materials The visual attractiveness and overall ambiance of the interior space can significantly impact a customer’s decision to purchase a property
Competitive Pricing: The price of a property compared to the market value is a critical factor influencing customer decisions Customers are more likely to choose properties that offer good value for money or are reasonably priced compared to similar options in the market Developers who offer competitive pricing strategies often attract more potential buyers
Financing Options: The availability of financing options, such as installment plans or flexible payment schedules, can greatly influence purchase decisions Many customers prefer properties that offer convenient payment options, allowing them to manage their finances effectively while acquiring their desired real estate assets
Legal Assurance: Ensuring legal compliance and proper documentation is crucial for customers when making real estate purchases Customers want assurance that the property they are buying has clear legal ownership and all necessary permits and approvals from relevant authorities Trustworthy developers who prioritize legal compliance can instill confidence in potential buyers
Reputation of Developer and Construction Contractor: The reputation of the developer and construction contractor involved in a real estate project can significantly impact purchase decisions Customers often consider the track record and credibility of the developer and contractor before investing in a property A reputable developer with a history of delivering high-quality projects can attract more customers.
THEORY - METHODS ofDATA MINING Ă.Ặ 2 c2 seẰ, 6
Data ClassIfication c1 2112112112121 112121111 111111111211 111 11181 1H Họ 6
Classification is a form of data analysis that extracts models describing important data classes Such models, called classifiers, predict categorical (discrete, unordered) class labels To divide bank loan applications into safe and dangerous categories, for instance, we may create a classification model Our comprehension of the facts as a whole may improve with the use of such analysis Researchers in machine learning, statistics, and pattern recognition have put forth many classification techniques Given a minimal amount of data, the majority of algorithms are memory-resident Based on earlier research, data mining algorithms have been developed recently that can handle enormous volumes of disk-resident data using scalable categorization and prediction methods
There are many uses for classification, such as in production, target marketing, fraud detection, performance prediction, and medicine
2.1.2 How does classification technique work? a) Inthe first step: Building the Classification Model
In this initial phase, the focus is on constructing a robust classification model leveraging historical or previous data This stage encompasses several key actions: Data Collection and Preparation: Gathering a comprehensive dataset that includes both input features and their corresponding labeled classes or categories This data is preprocessed to handle missing values, normalize features, and ensure its suitability for model training
Model Training: Utilizing various classification algorithms such as Decision Trees, Support Vector Machines (SVM), Logistic Regression, or Neural Networks, among others The selected algorithm learns from the provided dataset, establishing patterns and relationships between input features and their associated classes
Evaluation: Assessing the trained model's performance using validation techniques like cross-validation or by splitting the dataset into training and testing subsets This step ensures that the model has learned effectively without overfitting to the training data oe Classification algorithm name age income loan_decision
Sandy Jones youth low risky
Bill Lee youth low risky
Caroline Fox middle_aged high safe
Claire Phips senior medium safe
Joe Smith middle_aged high safe
IF age = youth THEN loan_decision = risky
IF income = high THEN loan_decision = safe
IF age = middle_aged AND income = low THEN loan_decision = risky (a) b) Inthe second step: Model Validation and Application
The second step involves evaluating the model's accuracy and readiness for real- world application: e Accuracy Assessment: Checking the model's performance metrics, such as accuracy, precision, recall, or Fl-score, to determine its effectiveness in correctly classifying instances This assessment ensures that the model generalizes well to new, unseen data e Application to New Data: If the model demonstrates acceptable accuracy and reliability during evaluation, it is deployed to classify new or incoming data The validated model becomes a predictive tool, assigning classes or categories to new instances based on the learned patterns from the training data
Classification rules ae One name age mcome loan_decision
Juan Bello — senior low safe
Sylvia Crest middle_aged low risky
Anne Yee middle aged high safe
(John Henry, middle_aged, low) Loan decision? risky
This two-step approach ensures a systematic and thorough process in developing, evaluating, and deploying classification models It emphasizes the importance of not only constructing an accurate model but also validating its performance before utilizing it for real-time classification tasks, thereby ensuring reliable decision- making based on data-driven insights
2.1.3 The basic techniques for data classification such as a) Decision tree classifiers
Flow Chart Structure: Decision tree classifiers create a tree-like structure resembling a flowchart This hierarchical representation begins with a root node, branches out into intermediate nodes, and concludes with leaf nodes.
Support in Decision-Making: These classifiers assist in decision-making by systematically processing input features and arriving at a decision or prediction Each node in the tree represents a feature or attribute, guiding the classification process based on specific conditions
Visual Representation of Rules: The rules for classification are visually defined within this tree structure As data moves through the tree, it undergoes a series of binary decisions based on feature thresholds, ultimately leading to the assignment of a class label at the leaf nodes
Root Node: Represents the primary question used to split the dataset based on the most significant feature
Branch Nodes: Intermediate nodes that embody the decision-making process, branching based on feature conditions
Leaf Nodes: Terminal nodes that provide the final outcome or class label for a specific subset of data
Consider a financial institution utilizing a decision tree classifier to assess credit scores for loan applications The institution classifies credit scores into four categories:
Using a decision tree, explain the sequential process the institution might follow to determine the credit score ratings for loan applicants Elaborate on the hypothetical criteria that the decision tree could employ at each node to classify applicants into these categories Describe how this decision-making process aids in categorizing individuals applying for loans based on their creditworthiness. f 1 1
Under 30 30 - 50 Above 30 under 15m/month 15 - 20 under 15 15-20 20 - 30 15 - 20 30 - 50
In establishing classification criteria for credit scores, several defining rules are in place denoted as A, B, C, and D These rules act as guiding principles within a decision-making framework designed to assess creditworthiness based on specific demographic and financial factors For instance, within this classification system, age and income thresholds play a pivotal role in determining an individual's credit score category Should an applicant fall under the age of 30 and have an income less than 15 million, the classification rule would allocate them to the 'Bad' credit score category, signifying a potential risk associated with this particular demographic segment Continuing within this decision tree framework, individuals under the age of 30 but with an income falling within the range of 15 million to 20 million are classified under the 'Good' credit score category This delineation highlights a different risk profile, acknowledging a more favorable credit assessment for individuals in this income bracket within the specified age group
These rules underscore the granularity and specificity of criteria used in evaluating creditworthiness By establishing such parameters based on age and income bands, the decision tree classifier effectively segregates applicants into distinct credit score categories, aiding financial institutions in assessing risk and making informed lending decisions b) Bayesian classifiers
Bayesian classifiers are a fascinating subset of statistical classifiers that employ a probabilistic approach to make predictions about the class membership of data items These classifiers are rooted in Bayes’ theorem, a fundamental principle in probability theory, and offer insightful ways to estimate the likelihood of an item belonging to a particular class or category
- Statistical Foundation: Bayesian classifiers operate on the premise of statistical analysis and probability theory They leverage prior knowledge about the probability distribution of classes and features within a dataset to make predictions
Probabilistic Predictions: One of the distinct features of Bayesian classifiers is their ability to provide not just a categorical prediction but also a probability estimate Rather than simply assigning a class label, they calculate the likelihood or probability that a given data item belongs to each potential class This probabilistic approach allows for a more nuanced understanding of uncertainty and confidence in predictions
- The fundamental formula underlying Bayesian classifiers
It mathematically expresses the relationship between conditional probabilities The theorem is represented as Naive bayes classifier
Posterior Probability Predictor Prior Probability
P(c|X) = P(x, |c)x P(x, |c)x -x P(x, |¢)x P(c) e P(c\x) is the posterior probability of class (target) given predictor (attribute) e P(c) is the prior probability of class e P(x\c) is the likelihood which is the probability of predictor given class e P(x) is the prior probability of predictor \
Data aSSOCIAEIOII 0 2 2 1222011212111 1111111212111 1 11111111 11 11 1111111181111 Hay 13 "N1 Ã 13
Association rules are derived from itemsets, which are sets of one or more items present in the data These sets could contain various attributes or features For instance, in market basket analysis, items purchased together might form an itemset
Example Rule: Consider the association rule: (Age > 25, Career) -> (Salary > 20m) This rule suggests that individuals over the age of 25 with a certain career tend to have salaries exceeding 20 million, revealing an association between age, career, and salary range
Unearthing Hidden Patterns: Association rules are valuable for uncovering concealed or implicit patterns within vast datasets They reveal connections between seemingly unrelated items or attributes, providing insights into customer behavior, market trends, or associations in various domains
Scalable Pattern Discovery: One of the strengths of association rules lies in their ability to learn scalable methods for pattern discovery from massive datasets Algorithms like Apriori and FP-Growth efficiently mine frequent itemsets, identifying sets of data items strongly correlated to each other despite the dataset's size
Definition: Support measures the frequency or occurrence of a specific itemset or rule within the dataset It indicates how frequently a particular combination of items appears together in the dataset
Significance: Higher support values imply that the itemset or rule appears frequently in the dataset Items or itemsets with high support are considered more significant and can be considered for further analysis
Example: Fraction of transactions that contain the itemset : (Milk, Disapers) => Beer) The support is:
40% of the transactions 0.4 —> Shown that milk, diapers,
|T| and beer are sold together o(milk, diapers, beer) _ sup = 2.2.4 Confidence
Confidence measures the reliability or strength of the rule It signifies the likelihood that the consequent of a rule holds true when the antecedent is present
Example: Fraction of times items in Y appear in transactions that contain X : (Milk, Diapers) => (Beer) cank= o(milk, diapers)
2.2.5 Lift o(milk, diapers, beer) _ 67% of the purchases
7 ằ that contain milk and diapers, also contain beer
Lift 1s to measures how much the likelihood of buying Y increases after knowing that X is also purchased pỉ|X) _ p(%Y) _ conf(XơY)
How much offects X in the decision of buying Y e Lift= 1 implies no association between A and B e Luift> 1 indicates that the presence of A positively influences the presence of B e Lift potato Chip pois chip support milk = P(milk) = 3/5 = 0.6 T1 support potato chip = P(potato chip) = 2/5 =
0.4 support = P(milk & potato chip) = 2/5 = 0.4 confidence
= support (milk & potato chip)/support(milk)
= 0.67 potato chip) /P(Milk) lift = confidence /support(potato chip) = 0.67/0.40 = 1.67 We
[P(Milk & Potato chip)/P(milk)) /P(Potato chip)
PRACTICE .-.- S2 22 21222112121 12121 151 2E2111111111121111 28111281 rve 15
Biometric characteristics of each customer ứroUp : ‹s-: 15
Based on these characteristics, we can observe that this survey focuses on gathering information about the distribution according to gender, age, and marital status of a group of people It is possible to identify 5 distinctive groups based on the data: 3.1.1 Group 1 - Singles, Mainly Purchasing Real Estate for Residence: Group 1 in the survey data primarily focuses on individuals who are single, constituting a significant proportion of the total customers (54.8%) This demographic exhibits distinctive biometric characteristics, being unmarried and without significant family responsibilities The main purpose of property acquisition for this group is residence, accounting for 78.6% This group often demonstrates independence and high autonomy in home-buying decisions With a planned budget ranging from low to medium, they may concentrate on finding homes with convenient locations and amenities suited to a single lifestyle
The flexibility and adaptability of this group may drive the demand for stable Irving spaces that can easily be adjusted to changes in individual life circumstances Therefore, understanding this group can assist developers and real estate professionals in formulating effective strategies to meet their specific needs 3.1.2 Group 2 - Young Families, Newly Married, Investing for the Future: Group 2 In the survey data represents Young Families, Newly Married, comprising 26.2% of the total customers This is a particularly crucial demographic in the real estate sector as they are in the midst of the newlywed stage and are preparing to build a happy family life The primary purpose of property acquisition for this group is to invest in the future, which may involve generating income through renting or establishing a stable family home With potentially limited budgets and financial planning, young families often focus on finding homes with amenities that cater to family needs, such as proximity to educational centers and parks
3.1.3 Group 3 - Elderly Families with Children, Purchasing Real Estate for Savings or Investment:
Group 3, focusing on Elderly Families with Children and showing a tendency to purchase real estate (RE) for savings or investment (19%), represents a significant demographic in the real estate market This group, characterized by family and parental responsibilities, brings stability and financial management experience to the table The presence of a previous home raises the need to cut costs or invest to optimize family finances Their investment strategy focuses on selecting real estate with the potential for appreciation and ensuring long-term profitability Their choice of real estate is also influenced by preferences for amenities and security in the residential area
3.1.4 Group 4 - Affluent Families, Purchasing Multiple Real Estates for Investment:
Group 4, representing "Affluent Families, Purchasing Multiple Real Estates for Investment" in the survey data, is a significant segment in the real estate sector With a high income ratio of up to 7.2%, this group stands out with a strategy of acquiring multiple real estates for investment purposes (21.4%) This group often demonstrates diversification in their investment strategy, focusing on risk management and optimizing returns from a diverse real estate portfolio It can be expected that these affluent families have a close relationship with the financial market and utilize real estate as a strategic tool to achieve financial goals and contribute to international asset diversification
3.1.5 Group 5 - Professional Real Estate Investors
This group, comprising professional investors in the late stages of their careers, all of whom are parents and intend to purchase real estate (RE) primarily for investment and savings, accounts for a proportion of up to 11.9% This demonstrates the maturity and strategic approach in financial management and investment by this group, showcasing their understanding of the value of growth and asset preservation through real estate
Some key characteristics of Group 5 include innovation in asset management, the ability to assess prior partnerships with RE, and the desire to expand their investment portfolio It's likely that they already own homes and are seeking opportunities to expand their investments, especially in the real estate sector For
Group 5, real estate is not just an investment tool but also a means to preserve asset value and optimize returns over time They tend to choose RE projects with potential for appreciation and stability, reflecting a particular interest in safeguarding asset values during the post-retirement phase
To better understand this demographic, real estate businesses should focus on developing projects and flexible business strategies that cater to their specific needs This approach can provide attractive investment solutions and support them in managing and expanding their asset portfolios effectively.
Characteristics of Customers’ Real Estate Preferences .- 17 1 Purpose of Real Estate AcquIsition : 2: ccs c2 x se sssxes 17 2 Type of Real Estate Desired for Purchase -. -.-55555- 18 3 Property S1Z@ Q0 1211201121 121111 1211 1101111111101 111018 H1 19 3.3 FInancIal StrCfUTeS .- - - 2L 2C 22 2022011202211 151 15111511111 51 11111811811 1 ke 20 3.4 Financial Characteristics of Customers - - 5c 2-2 2212212122122 23 3.4.1 Planned Budget for Home Purchase by Customers with Monthly Ineome of20-30 Million VND 2L 1 2122111221211 1211211811 23 3.4.2 Planned Budget for Home Purchase of Customers with Monthly Incomes of 50-70 Million VND uuu cccccccccccccccccssensceseensteseetsetseensenieens 24 3.4.3 Planned Budget for Home Purchase of Customers with Monthly Ineomes Above 70 Milion VND c2 2c 121121 11112112121 18211 mre 25
3.2.1 Purpose of Real Estate Acquisition
Purpose of buying real estate Investment
According to the chart, it can be observed that the primary purpose of real estate acquisition for the research participants is for residential purposes, accounting for the majority at 78.6% The remaining portion is attributed to investment purposes, constituting 21.4% Within the residential category, the predominant group is Group
2 (young couples purchasing their first apartment for independent living), followed by Group | and some participants from Group 3 and Group 4 Additionally, the investment-oriented purpose is mainly associated with participants from Group 5 and Group 4
The reason for the higher percentage of real estate acquisition for residential purposes compared to investment purposes is that the majority of research participants fall into Groups 1, 2, and 3 When they have sufficient capital to purchase real estate, they prioritize acquiring properties to meet their housing and ltving needs After accumulating surplus capital, demonstrating good financial capabilities and high income, they then tend to consider additional investments According to the research data, most of the research participants have the financial capacity to purchase homes, but not all of them have surplus financial resources for further real estate investments Therefore, residential purposes remain the primary choice for these groups, with investment being a secondary consideration for those with greater financial capacity and resources
3.2.2 Type of Real Estate Desired for Purchase
Type of Real Estate Desired for Purchase
According to the data, it can be observed that the most desired type of property for research participants is condominiums, accounting for 52.4%, followed by standalone houses at 30.9%, and finally, purchasing land for self-construction (land) at 16.7% One of the significant reasons many people choose to buy condominiums is the potential cost savings compared to purchasing a standalone house Maintenance and repair costs for condominiums are often shared among residents, helping alleviate financial pressure for each individual family Additionally, living in urban areas can save commuting costs and provide daily conveniences
Condominium buildings typically offer amenities such as gyms, playgrounds, tennis courts, and are conveniently located near shopping centers, schools, and hospitals This brings convenience and time savings for residents Furthermore, buying a pre- built house is often cheaper than purchasing land and constructing a house from scratch This is because land acquisition costs, construction materials, labor, and other expenses are already factored into the house price Additionally, it saves time, as constructing a house from scratch can take several months to a year or more, depending on the complexity of the design and the availability of materials.
Among these, Group | and Group 2 comprise the majority when it comes to choosing condominiums as their residence This is because young individuals tend to prioritize time and effort savings when buying a home, avoiding concerns related to urban planning, permits, construction, or additional costs Additionally, meeting the needs of the younger demographic involves having a prime location with convenient transportation, either in the city center or near densely populated and developing residential areas Condominiums offer various upscale and modern amenities such as shopping centers, restaurants, cafes, playgrounds, hospitals, schools, swimming pools, gyms, BBQ areas, and more
On the other hand, standalone houses are typically preferred by research participants in Group 3 and some in Group 4 and 5 These groups often prioritize a long-term perspective Additionally, standalone houses are viewed as having a high appreciation potential, can be passed down to future generations, or sold to upgrade to another house when needed
Finally, purchasing land for self-construction is the most common among participants in Group 5 Land is considered the highest-priced type of real estate compared to condominiums and standalone houses It offers long-term ownership rights, high appreciation potential, the flexibility to build according to personal preferences, and is exempt from management fees, services, and parking fees Therefore, it is seen as a prudent choice for investors seeking to reap long-term benefits
The desired property size for research participants often falls within the range of 50- 80m2 (40.5%) and 80-110m2 (26.2%), with the majority leaning towards smaller sizes Following this, we have the range of 110-150m2 (21.4%) and over 150m2 (11.9%) It is evident that research participants prioritize purchasing homes with smaller sizes This is likely because buying a smaller-sized home often entails lower costs compared to purchasing a larger one, enabling buyers to save money and manage repayments more easily Additionally, smaller homes are generally easier to manage, requiring less time for cleaning and maintenance
With the modern living trend, many young individuals prefer to live in apartments or houses with smaller sizes to maximize space utilization, amenities, and cost savings Therefore, they often choose condominiums with sizes ranging from 50- 80m2, fitting their needs and financial capabilities Hence, Group 1, consisting of young singles, and Group 2, young families, mostly prioritize sizes within the range of 50-80m2 On the other hand, for families with children (Group 3), they may prefer larger sizes to accommodate household activities and provide ample space for family members
Considering the above, Group 4 tends to choose sizes ranging from 80-110m2 or 110-150m2 to meet the needs of each family Finally, the choices regarding property size for Group 5 are quite complex, depending on investment trends, market demands, and individual investment goals Therefore, they are evenly distributed across all four size categories
@ Existing savings ® Borrow in full
Observing the chart, it can be noted that the majority of the research subjects tend to accumulate a portion of the funds, while the remaining part is financed through loans (80.9%) Additionally, some individuals also opt for a combination of existing savings and loans The stability in finances and seizing opportunities quickly are the primary reasons why many people choose to accumulate a portion and finance the rest This is driven by the desire to secure a property that meets their criteria as soon as it becomes available in the market However, lacking sufficient savings to make the purchase outright often leads individuals to consider the unavoidable option of taking a bank loan The emergence of additional incidental costs during the home buying or construction process also influences the decision to resort to bank loans among the research subjects Moreover, this trend aims to capitalize on opportunities and prospects when selecting a desirable home or a rare investment opportunity, contributing long-term value for individuals, families, and future investors However, when choosing this trend, it is crucial to understand financial principles, estimate the payback period, to avoid unfortunate risks and achieve the most efficient profit values
The trend of accumulating a portion and financing the rest is prevalent across all five groups of subjects, with the highest representation in Group 1 (successful, independent, high-income young individuals) and Group 5 (professional real estate investors) In Group 1, young individuals often lack a significant amount of money to purchase a property outright, but due to financial independence, stable jobs, and incomes, they possess stable financial resources to repay the borrowed amount when buying a house Additionally, in Group 5 (professional real estate investors), they tend to seize the opportunity to buy property based on future potential, land values, etc To capitalize on this, they need strong and timely financial resources Therefore, borrowing from a bank to choose investments is a wise choice for investors, aiming to bring economic benefits to this group © Bank loan
@ borrow from family and relatives
@ credit institution loan" borrow from family and relatives
Based on the observed data, it can be seen that bank loans are the primary source of financial borrowing for the majority of the research subjects (73.8%), followed by loans from family and relatives (21.4%), and lastly, loans from credit institutions (7.8%) Bank loans are evenly distributed across all five groups of research subjects, with Groups | (successful, independent, high-income young individuals) and 5 (professional real estate investors) having the highest proportions These trends are influenced by factors such as trustworthiness and reputation Banks are reputable financial institutions with a long history of lending to customers, providing borrowers with a sense of security Additionally, banks offer various repayment options based on individual borrower characteristics, providing flexibility Moreover, banks often present competitive interest rates on their loans, helping borrowers save money on interest payments throughout the loan period This can be particularly crucial for those requiring a substantial loan amount In addition, many individuals tend to borrow from family and relatives rather than credit institutions due to concerns about risk, fraud, credibility, and high interest rates associated with credit institutions
Group | Purpose Type of Real Property | Financial form] Financial
Estate Desired for size loan source
Group Living Apartment 50-80 Accumulate in Bank
Group Living Apartment 50-80 Accumulate in| Family,
Group Living Real estate 50-80 / | Accumulate in| Family,
Group | Investment Real estate 80-110; | Accumulate in Family,
Group | Investment Land All Accumulate in Bank
From 1-2 From 2-3 Under 1 Over3 billion billion billion billion
3.4.1 Planned Budget for Home Purchase by Customers with Monthly Income of 20-30 Million VND
In the context of an increasingly diverse real estate market, understanding and meeting the needs of each target group become crucial Group 1, consisting of singles whose main purpose of purchasing real estate is for residence, with a stable income ranging from 20-30 million VND per month, is a particularly noteworthy group in the real estate market This group, characterized by independence and a primary goal of finding a home to live in, presents both challenges and opportunities According to detailed analysis, approximately 30% of the total 20 individuals in the group are interested in homes priced under 1 billion VND The number of interested individuals increases to 45% when the price limit is between 1-2 billion VND, while 15% have a budget of 2-3 billion VND, and 10% are willing to make a purchasing decision for homes priced over 3 billion VND