Hiren Patel IU2241230273​​ ​ ​ ​ 6th CSE

Data Preparation
& Analysis
(CS0602)
Practical File

Hiren Patel
B.Tech CSE
Semester 6
IU2241230273

Data Preparation & Analysis 1

Hiren Patel IU2241230273​​ ​ ​ ​ 6th CSE

PRACTICAL - 1
AIM:
Study about Advantages & Disadvantages of Data Preparation & Analysis.

Data Preparation

Advantages:

1.​ Data Cleaning:

○​ Improves Data Quality: Removing duplicates, handling missing values, and
detecting outliers lead to more accurate and reliable data.
○​ Enhances Analysis: Clean data ensures that the analysis is based on true
representations, resulting in better insights.
2.​ Data Transformation:
○​ Consistency: Normalization/standardization ensures that data is on a common
scale, making it easier to compare.
○​ Usability: Encoding categorical variables makes them usable in machine learning
algorithms.
3.​ Data Integration:
○​ Comprehensive Data: Combining datasets provides a more complete picture.
○​ Multimodal Analysis: Integrating different types of data can lead to richer
insights.
4.​ Data Reduction:
○​ Efficiency: Reducing the dimensionality or size of the data can speed up the
analysis process.
○​ Focus on Key Features: Helps in focusing on the most important features that
have the most impact.

Disadvantages:

1.​ Time-Consuming: The process of cleaning, transforming, integrating, and reducing data
can take a significant amount of time and effort.
2.​ Complexity: Requires expertise and careful handling to avoid introducing errors or bias.
3.​ Resource Intensive: Involves the use of various tools and skilled personnel, which can be
costly.

Data Analysis

Advantages:

1.​ Descriptive Statistics:

Data Preparation & Analysis 2

Hiren Patel IU2241230273​​ ​ ​ ​ 6th CSE

○​ Summarizes Data: Provides a quick summary of the dataset's main

characteristics.
○​ Easy to Interpret: Simple measures like mean, median, and mode are easy to
understand.
2.​ Exploratory Data Analysis (EDA):
○​ Identifies Patterns: Helps in discovering hidden patterns and trends.
○​ Visual Insights: Data visualization makes it easier to understand complex data.
3.​ Inferential Statistics:
○​ Makes Predictions: Allows for making inferences about a population based on a
sample.
○​ Hypothesis Testing: Helps in determining the statistical significance of findings.
4.​ Predictive Analytics:
○​ Forecasting: Provides predictions about future events based on historical data.
○​ Decision Making: Supports better decision-making by providing actionable
insights.
5.​ Clustering:
○​ Group Identification: Identifies groups or clusters within data.
○​ Market Segmentation: Useful for segmenting markets or customer bases.
6.​ Association Rule Learning:
○​ Discover Relationships: Identifies associations between different variables.
○​ Market Basket Analysis: Helps in understanding purchase patterns.

Disadvantages:

1.​ Complexity: Analysis techniques can be complex and require specialized knowledge.
2.​ Data Quality Dependency: Results are heavily dependent on the quality of the input
data.
3.​ Resource Intensive: Requires computational resources and expertise to perform
effectively.
4.​ Potential for Misinterpretation: Incorrect analysis or misinterpretation of results can
lead to wrong conclusions.

Data Preparation & Analysis 3

Hiren Patel IU2241230273​​ ​ ​ ​ 6th CSE

PRACTICAL - 2
AIM:
Study about Different data analytical tools.

Tableau

Overview:

Tableau is a Business Intelligence (BI) and data visualization tool that helps users analyze and
present data through interactive dashboards and reports.

Key Features:

●​ Tableau Prep for data cleaning and transformation.

●​ Drag-and-drop interface for data visualization.
●​ Connects to multiple data sources (databases, cloud, Excel, etc.).
●​ Real-time data analytics with live or extracted data.

Applications:

●​ Business intelligence and decision-making.

●​ Market trend analysis.
●​ Customer insights and reporting.

Example Use Case:

●​ Retail: Analyzing sales trends and customer demographics.

●​ Finance: Monitoring real-time financial transactions and risk assessment.
●​ Healthcare: Visualizing patient data for better diagnosis and treatment planning.

2. Power BI

Overview:

Power BI is a Microsoft-powered BI tool used for data visualization, reporting, and analytics,
particularly in enterprise environments.

Key Features:

●​ Power Query for data extraction, transformation, and loading (ETL).

Data Preparation & Analysis 4

Hiren Patel IU2241230273​​ ​ ​ ​ 6th CSE

●​ Integrates seamlessly with Excel, SQL Server, and Azure.

●​ AI-driven insights and automation capabilities.
●​ Real-time dashboard updates.

Applications:

●​ Enterprise reporting and dashboards.

●​ Data-driven business decision-making.
●​ Financial planning and forecasting.

Example Use Case:

●​ Corporate Performance: Analyzing company-wide KPIs and operational metrics.

●​ Sales & Marketing: Tracking customer engagement and campaign performance.
●​ Supply Chain: Monitoring inventory and logistics efficiency.

3. Hadoop

Overview

Hadoop is an open-source framework for distributed data storage and processing, primarily used
for handling big data.

Key Features:

●​ HDFS (Hadoop Distributed File System) for storing large datasets.

●​ MapReduce for parallel processing of massive data.
●​ Works with Hive and Pig for querying and transforming data.
●​ Supports structured, semi-structured, and unstructured data.

Applications:

●​ Big data storage and processing.

●​ Large-scale data analytics.
●​ Enterprise data warehousing.

Example Use Case:

●​ E-commerce: Processing millions of customer transactions in real-time.

●​ Healthcare: Storing and analyzing large volumes of patient health records.
●​ Social Media: Analyzing user-generated data across platforms.

Data Preparation & Analysis 5

Hiren Patel IU2241230273​​ ​ ​ ​ 6th CSE

4. Apache Spark

Overview:

Apache Spark is an in-memory big data processing engine designed for real-time and batch data
analytics.

Key Features:

●​ Spark SQL for querying structured data.

●​ Spark Streaming for real-time data processing.
●​ MLlib for machine learning and data science applications.
●​ GraphX for graph processing.

Applications:

●​ Real-time big data analytics.

●​ AI and machine learning pipelines.
●​ Streaming data processing.

Example Use Case:

●​ Banking & Finance: Detecting fraud in real-time transactions.

●​ Telecommunications: Analyzing customer call data for better service.
●​ IoT (Internet of Things): Processing sensor data in real-time.

5. TensorFlow

Overview:

TensorFlow is an open-source machine learning and AI framework used for building and training
deep learning models.

Key Features:

●​ TensorFlow Data (TFData) for scalable data preparation.

●​ Supports deep learning models (CNNs, RNNs, NLP).
●​ TensorFlow Lite for mobile and embedded AI applications.
●​ Cloud-based model training with TensorFlow Serving.

Applications:

Data Preparation & Analysis 6

Hiren Patel IU2241230273​​ ​ ​ ​ 6th CSE

●​ Artificial intelligence and deep learning.

●​ Natural language processing (NLP).
●​ Image and speech recognition.

Example Use Case:

●​ Healthcare: Predicting diseases using AI-powered diagnostics.

●​ Autonomous Vehicles: AI-driven object detection and path planning.
●​ Finance: Fraud detection using deep learning models.

Data Preparation & Analysis 7