Data Science

The foundational methodology for Data Science is often represented by a cycle or flow diagram.
One widely used diagram is the CRISP-DM (Cross-Industry Standard Process for Data Mining) model.
It includes the following steps:
1. Business Understanding
2. Data Understanding
3. Data Preparation
4. Modeling
5. Evaluation
6. Deployment
A simple way to draw it is as a circular flowchart showing how each stage connects in a loop.
This shows that Data Science is an iterative process with continuous refinement.
Students should draw this diagram in their notebooks with arrows indicating the flow between stages.

Step 1: Data Collection.
Large volumes of structured and unstructured data are collected from multiple sources such as social media, sensors, and databases.

Step 2: Data Storage.
The collected data is stored using technologies like Hadoop or cloud storage systems to handle large-scale data efficiently.

Step 3: Data Processing.
Data is cleaned, organized, and processed using tools like MapReduce or Spark to extract useful information.

Step 4: Data Analysis and Visualization.
Processed data is analyzed to identify patterns and insights, and results are presented using charts, dashboards, and reports.

Step 1: Define Artificial Neural Network (ANN).
An Artificial Neural Network (ANN) is a computational model inspired by the human brain, consisting of interconnected nodes (neurons) that process information.

Step 2: Explain Input Layer.
The input layer is the first layer of the network that receives raw data from the external environment. Each neuron in this layer represents a feature of the input data.

Step 3: Explain Hidden Layer.
The hidden layer(s) lie between input and output layers and perform complex computations. These layers process the input using activation functions and extract patterns or relationships.

Step 4: Explain Output Layer.
The output layer produces the final result of the network, such as classification or prediction, based on the processed data.

Step 5: Role of Weights.
Weights are numerical values assigned to connections between neurons. They determine the importance of input signals. During training, these weights are adjusted to minimize error and improve accuracy.

Step 1: Understanding neural network flow.
The question describes the process where input passes through layers, activations are calculated, and output is generated.

Step 2: Differentiating processes.
• (A) Back Propagation: Used to update weights based on error. • (B) Deep Learning: A broader concept, not a specific step. • (C) Forward Propagation: Correct. Input flows forward through the network to produce output. • (D) Optimization: Refers to improving model parameters.
Step 3: Conclusion.
Thus, the described process is Forward Propagation.
Final Answer: Forward Propagation.

Step 1: Understanding the scenario.
The focus is on ensuring data is accurate, consistent, reliable, and trustworthy before analysis.

Step 2: Big Data characteristics.
• (A) Volume: Amount of data. • (B) Velocity: Speed of data generation. • (C) Variety: Different types of data. • (D) Veracity: Quality, accuracy, and reliability of data.
Step 3: Conclusion.
Since the scenario emphasizes data quality and trustworthiness, it represents Veracity.
Final Answer: Veracity.

Step 1: Understanding the scenario.
Question me fraud detection ki baat ho rahi hai jahan unusual transactions ko identify kiya jata hai jo normal pattern se different hote hain.

Step 2: Concept of anomaly detection.
Anomaly detection ek technique hai jo dataset me unusual ya abnormal patterns ko identify karti hai. Yeh fraud detection, security aur monitoring systems me widely use hoti hai.

Step 3: Analysis of options.
• (A) Clustering: Similar data ko group karta hai, anomalies detect nahi karta directly.
• (B) Recommendation: Suggestions dene ke liye use hota hai.
• (C) Regression: Prediction ke liye use hota hai.
• (D) Anomaly Detection: Correct. Yeh abnormal patterns detect karta hai.

Step 4: Conclusion.
Isliye fraud detection ka process anomaly detection se related hai.
Final Answer: Anomaly Detection.

Step 1: Understanding the scenario.
The company is experiencing a decline in sales and is trying to find the reasons behind it by analyzing various factors such as customer behavior, traffic, and return rates.

Step 2: Understanding types of analytics.
• Descriptive Analytics: Tells what has happened in the past. • Diagnostic Analytics: Explains why something happened. • Predictive Analytics: Forecasts future trends. • Prescriptive Analytics: Suggests actions to take.
Step 3: Matching with the scenario.
Since the team is investigating the cause of the decline, they are focusing on understanding why it happened, which is the main goal of diagnostic analytics.

Step 4: Conclusion.
Thus, the correct type of analytics being used is Diagnostic Analytics.
Final Answer: Diagnostic Analytics.

Step 1: Understanding the data type.
The question involves audio, images, and videos, which do not follow a fixed structure like tables or databases.

Step 2: Types of Big Data.
• (A) Structured Data: Organized in tables with rows and columns. • (B) Semi-Structured Data: Has partial structure (like XML or JSON). • (C) Unstructured Data: Correct. Includes audio, images, and videos without predefined format. • (D) Filter Data: Not a recognized category of Big Data.
Step 3: Conclusion.
Hence, the analyst is dealing with unstructured data.
Final Answer: Unstructured Data.