Scribd
Upload
3 views18 pages

Chapter 2

The document discusses the contrast between intuition-based real estate valuation and data-driven models, emphasizing the importance of using structured data analysis for more accurate predictions. It introduces decision trees as a powerful tool for modeling and visualization in real estate, highlighting their simplicity, interpretability, and ability to handle complex data interactions. Additionally, it provides a checklist for machine learning projects, outlining essential steps for successful implementation.

Uploaded by

Youssef Mrakchi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views18 pages

Chapter 2

The document discusses the contrast between intuition-based real estate valuation and data-driven models, emphasizing the importance of using structured data analysis for more accurate predictions. It introduces decision trees as a powerful tool for modeling and visualization in real estate, highlighting their simplicity, interpretability, and ability to handle complex data interactions. Additionally, it provides a checklist for machine learning projects, outlining essential steps for successful implementation.

Uploaded by

Youssef Mrakchi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

1

Real Estate
Valuation

[email protected] NEXT
2

Cousin's Success
Story
Cousin's Intuition and Success
Your cousin has made significant wealth through real estate
speculation, relying mainly on intuition. Through keen
observation of price patterns in houses he views, he
believes he can predict future values effectively. This
experience raises questions about the potential for data
analysis to improve his predictions. Leveraging your
expertise in data science could provide a more structured
and accurate approach to valuation.

NEXT
3

Intuition vs. Models

Intuition-Based Predictions Data-Driven Models


Using intuition can lead to quick Data-driven models utilize historical
decisions but may lack consistency. It data to uncover patterns and trends
often relies on emotional judgment that inform predictions. They offer
and personal experience, which can consistency and objectivity which are
differ widely among individuals. often missing with intuition.
Furthermore, past experiences might Moreover, these models can adapt to
not always reflect current market changes in the market by
dynamics, leading to errors in incorporating new data, enhancing
valuation. Although intuition has its accuracy over time. Machines learn
place, it's essential to consider its from vast datasets, providing insights
limitations. that intuition might overlook.

NEXT
4

Decision Tree
Basics
Introduction to Decision Trees
Decision trees are simple yet powerful tools for modeling
and visualization in data science. They partition data into
branches to make clear, actionable predictions based on
various features of the houses, like size and location. This
model operates through a tree-like structure where each
node represents a decision based on input features. Their
ease of understanding makes them an excellent starting
point for incorporating machine learning into real estate
valuation.

NEXT
5

Decision tree learning is a supervised learning approach


used in statistics, data mining and machine learning.
A decision tree is a type of supervised learning
algorithm that is commonly used in machine learning
to model and predict outcomes based on input data.

NEXT
6

Advantages of Simplicity and Interpretability

Decision Trees
One of the main advantages of decision trees is their clear
structure, making it easy to explain predictions to stakeholders.
Their visual representation allows quick interpretation, helping
users understand the decision-making process. Furthermore,
they don't require extensive data preprocessing, which
simplifies the modeling process. This accessibility makes them
an attractive option for newcomers to machine learning.

Handling Non-Linear Data


Decision trees effectively manage both categorical and
continuous data. Their ability to capture complex
interactions between features enhances prediction quality.
Additionally, trees can be easily modified to address
overfitting, ensuring that models remain robust in various
scenarios. This adaptability is crucial for accurately
valuating diverse properties in fluctuating markets.

NEXT
7

NEXT
8

It divides houses into only two categories. The predicted price


for any house under consideration is the historical average
price of houses in the same category.
We use data to decide how to break the houses into two
groups, and then again to determine the predicted price in
each group. This step of capturing patterns from data is called
fitting or training the model. The data used to fit the model is
called the training data.
The details of how the model is fit (e.g. how to split up the
data) is complex enough that we will save it for later. After the
model has been fit, you can apply it to new data to predict
prices of additional homes.

NEXT
9

Improving the Decision Tree

Which of the following two decision trees is more likely to


result from fitting the real estate training data?

NEXT
10

Improving the Decision Tree: 2

The decision tree on the left (Decision Tree 1) probably makes


more sense, because it captures the reality that houses with
more bedrooms tend to sell at higher prices than houses with
fewer bedrooms. The biggest shortcoming of this model is that
it doesn't capture most factors affecting home price, like
number of bathrooms, lot size, location, etc.
You can capture more factors using a tree that has more
"splits." These are called "deeper" trees. A decision tree that
also considers the total size of each house's lot might look like
this:

NEXT
11

NEXT
12

DataSet

You predict the price of any house by tracing


through the decision tree, always picking the
path corresponding to that house's
characteristics. The predicted price for the house
is at the bottom of the tree. The point at the
bottom where we make a prediction is called a
leaf.
The splits and values at the leaves will be
determined by the data, so it's time for you to
check out the data you will be working with.

https://www.kaggle.com/datasets/dansbecker/melbourne-
housing-snapshot

NEXT
13

NEXT
14

Model Comparison Table


M o d el Typ e D escrip tio n A ccuracy

Decision Tree Simple and interpretable model 75%

Random Forest Ensemble of trees for stronger predictions 85%

Linear Regression Basic linear models for continuous output 70%

Support Vector Machine Effective for high-dimensional spaces 80%

Neural Networks Deep learning for complex patterns 90%

NEXT
15

This checklist can guide you through your Machine Learning projects. There are
eight main steps:
1. Frame the problem and look at the big picture.
2. Get the data.
3. Explore the data to gain insights.
4. Prepare the data to better expose the underlying data patterns to Machine Learn‐
ing algorithms.
5. Explore many different models and short-list the best ones.
6. Fine-tune your models and combine them into a great solution.
7. Present your solution.
8. Launch, monitor, and maintain your system.
Obviously, you should feel free to adapt this checklist to your needs.

NEXT
16

This Photo by Unknown Author is licensed under CC BY-SA

NEXT
17

NEXT
18

Complete the in class practice in canvas

NEXT

You might also like