GET SET PY - Problem statements
PS01 - “SoundScribe" - Audio Genre Evolution Analyzer
Problem Statement:
In the rapidly evolving music industry, the popularity of tracks often dictates an artist's reach and
success. As music streaming platforms continue to dominate, understanding what factors
influence a track's popularity is crucial for artists, producers, and industry professionals. The
goal of this task is to develop a linear regression model that predicts the popularity of a music
track based on various audio and track-related features.
Using the given dataset, which includes information about track ID, artist name, album name,
track name, and key features such as danceability, energy, and duration, your task is to build
a predictive model. The dataset also contains attributes like whether the track is explicit and its
popularity score.
[Link]
PS02 - Maximizing Returns: Analyzing the Impact of Advertising
Investments on Sales Revenue
Problem Statement:
You are provided with a dataset that captures the investment in various advertising channels
and the resulting revenue from sales campaigns. The goal of this problem is to predict the
revenue generated from sales based on the investments made in three different advertising
mediums: TV, Radio, and Newspaper.
Dataset Details:
● TV: Investment in TV advertising campaigns (in thousands of dollars).
● Radio: Investment in radio advertising campaigns (in thousands of dollars).
● Newspaper: Investment in newspaper advertising campaigns (in thousands of dollars).
● Sales: Revenue generated from sales campaigns (in thousands of dollars).
Your task is to build a linear regression model that accurately predicts Sales based on the
advertising expenditures in TV, Radio, and Newspaper. This model will help businesses
optimize their advertising spending by understanding how each advertising medium contributes
to the overall sales revenue.
[Link]
PS03 - Effect of Fertilizer and Water on Crop Yields
Problem Statement
The research titled "Effect of Fertilizer and Water on Crop Yields" aims to investigate the
influence of varying amounts and types of fertilizers, along with different irrigation levels, on the
yield of various crops. The dataset includes key numerical features such as the amount of
fertilizer applied (in kilograms), water usage (in liters), and the resultant crop yield (in tons per
hectare), alongside environmental factors such as temperature and humidity. By employing
linear regression techniques, this study will analyze the relationships between these inputs and
crop yields, aiming to provide insights that can enhance agricultural practices and optimize
resource usage in farming.
Objective
The primary objective of this study is to quantitatively assess how different fertilizer types and
quantities, in conjunction with water availability, impact crop yields. By leveraging linear
regression modeling, the analysis seeks to identify significant predictors of yield, enabling
farmers and agricultural stakeholders to make data-driven decisions regarding the optimal
application of fertilizers and irrigation practices to maximize crop production while ensuring
sustainable farming methods.
[Link]
PS04 - Health & Lifestyle Analytics Dashboard Theme: Personal Health
Analysis
Problem Statement Description
The study titled "Health & Lifestyle Analytics Dashboard Theme: Personal Health Analysis" aims
to analyze the interrelationships between various health metrics and lifestyle choices to
understand their impact on overall personal health. Using a dataset that encompasses a range
of factors—including physical activity levels, dietary habits, sleep quality, body measurements,
and medical history—this analysis employs linear regression techniques to model and predict
health outcomes based on these lifestyle variables. The goal is to create an interactive analytics
dashboard that visualizes these relationships, providing users with insights into how their daily
habits may influence their health status and guiding them towards healthier lifestyle choices.
Objective
The primary objective of this analysis is to identify and quantify the impact of specific lifestyle
factors—such as exercise frequency, nutrition, and sleep patterns—on key health
indicators,including body mass index (BMI), cholesterol levels, and blood pressure. By utilizing
linear regression, the project aims to develop predictive models that can help individuals
understand how modifications in their lifestyle choices can lead to improvements in their health
metrics, ultimately fostering a proactive approach to personal health management.
[Link]
PS05 - Optimizing Customer Satisfaction: Predicting Service Levels with
Call Center Performance Metrics
Problem Statement:
In today's competitive market, delivering exceptional customer service is crucial for retaining
customers and driving business growth. Call centers play a pivotal role in shaping customer
experience, and understanding how various performance metrics affect overall satisfaction is
key to improving service quality.
In this project, the task is to develop a linear regression model to predict Customer Satisfaction
(measured as Service Level) using a dataset of call center metrics. The dataset includes
important performance indicators such as:
Dataset Details:
● Answer Rate: The percentage of calls successfully answered by agents.
● Answer Speed: The average time it takes for an agent to respond to a customer call.
● Talk Duration: The duration of conversations between the customer and the call center
agent.
● Wait Time: The average waiting period customers experience before speaking with an
agent.
The goal is to identify how these factors influence overall customer satisfaction and predict the
Service Level based on these operational metrics. By building an accurate predictive model, the
management will be able to pinpoint areas of improvement, optimize resources, and ultimately
enhance the customer experience.
[Link]
PSO6 - Predicting House Prices
Problem Statement:
The real estate market is influenced by various factors that affect the pricing of houses.
Understanding these factors is crucial for homebuyers, sellers, and real estate agents to make
informed decisions. In this project, the goal is to develop a linear regression model to predict
House Prices based on essential property features.
The dataset includes the following input features (independent variables):
● Area : The total size of the property.
● Number of Bedrooms : The total number of bedrooms in the house.
● Main Road: A binary variable indicating whether the house is located near a main road
(Yes, No).
● Number of Stories : The total number of floors or stories in the house.
The objective is to create a predictive model that accurately estimates house prices based on
the property’s size, number of bedrooms, proximity to a main road, and number of stories. This
model will help individuals and real estate professionals assess property values more effectively,
using the identified key factors driving house prices.
[Link]
PS07 - Insurance Charges Predictor
Problem Statement:
In this project, your goal is to build a predictive model that estimates the insurance charges for
individuals based on their demographic and health-related data. The dataset contains various
features such as the age of the insured, gender, body mass index (BMI), number of children,
smoking habits, and the region where the individual resides. The target variable, "charges,"
represents the premium paid for insurance coverage.
Objective:
1. Build a Linear Regression Model: Use the provided features to predict the insurance
charges using linear regression. Evaluate the performance of the model and provide insights
into which factors (such as age, BMwI, or smoking status) have the most significant impact on
the premium amount.
2. Develop a User Interface with Streamlit: Participants are required to create an interactive
UI using Streamlit, where users can input their demographic details (age, gender, BMI, number
of children, smoking status, region) and obtain a prediction of their estimated insurance
charges. The predicted charges will be calculated by the model and displayed in the UI.
3. Visualize Predictions and Insights: Display the predicted insurance premium, and
optionally provide visualizations such as the correlation between input features and predicted
charges, using interactive charts and graphs.
[Link]
PS08 - Predicting Employee Productivity Based on Work Environment
Factors
Problem Statement:
In this project, you are tasked with building a predictive model to estimate employee productivity
based on various work environment factors. The dataset includes variables such as stress
levels, working hours, and the number of leaves taken. The target variable, "productivity_rate,"
represents the employee's productivity level on a numerical scale.
Objective:
The goal is to create a linear regression model that predicts the productivity rate of an employee
based on the provided work environment factors. After building the model, students are required
to create a frontend using Streamlit, where users can input the relevant features (stress level,
working hours, and leaves taken) to predict employee productivity. This user interface will allow
for real-time predictions and serve as a tool for HR departments or managers to estimate how
changes in work conditions may impact productivity.
Key Tasks:
1. Build a linear regression model using the dataset.
2. Develop an interactive frontend using Streamlit for data input and model prediction.
3. Analyze the impact of different work environment factors on employee productivity and
provide insights based on the model's output.
Dataset:
[Link]
PS09 - Unlocking Growth: Predicting Company Expansion Amid Market
Dynamics
Problem Statement: In an ever-evolving business environment, companies must navigate
complex market conditions, competitor strategies, and consumer behavior to sustain and drive
growth. This dataset offers insights into the factors influencing a company's growth rate,
allowing for an in-depth analysis of how external and internal variables impact business
expansion.
Your goal is to develop a linear regression model that predicts the Company Growth Rate
based on various indicators, including market performance, inflation trends, competitor growth
rates, consumer spending habits, and the company’s advertising budget. By interpreting these
relationships, your analysis can aid the company in making data-driven decisions to allocate
resources effectively and enhance growth strategies.
Dataset:
[Link]
Should have in your project
1. OLS stats using statsmodels
2. Minimum of 2 visualizations
3. Understand 5 failing points in Linear regression