1. Explain Text Analytics and Text Mining.

Text Analytics refers to the process of analyzing unstructured text data

to extract useful information, identify patterns, trends, and relationships,
and support decision-making.
Key Features of Text Analytics
• Converts unstructured text into structured data
• Uses Natural Language Processing (NLP) techniques
• Identifies keywords, entities, topics, and sentiments
• Helps in business intelligence and strategic decisions
Applications of Text Analytics
• Customer feedback analysis
• Social media monitoring
• Email and document classification
• Fraud detection and risk analysis
• Healthcare and legal document analysis

Text Mining
Text Mining is a subset of data mining that focuses on discovering
hidden patterns and knowledge from large collections of text using
statistical, machine learning, and linguistic techniques.
Key Features of Text Mining
• Extracts implicit knowledge from text
• Uses data mining and machine learning algorithms
• Works on large text corpora (document collections)
• Emphasizes pattern discovery rather than simple analysis
Applications of Text Mining
• Market basket analysis from customer reviews
• Research literature analysis
• Spam filtering
• Trend detection in news articles
• Biomedical text analysis

Aspect Text Analytics Text Mining

Focus Understanding and Knowledge discovery
interpretation of text and pattern extraction
Approach NLP-based analysis Data mining and ML-
based
 Output Structured insights, Patterns, rules,
summaries predictions
Usage Business reporting and Advanced predictive
decision support analysis

2. Explain the Text Mining Process.”

The Text Mining Process refers to a systematic set of steps used to

extract useful information, patterns, and knowledge from large
volumes of unstructured text data such as documents, emails, web pages,
reviews, and social media content. Since textual data is unstructured, it
must be transformed into a structured format before analysis.

Steps in the Text Mining Process

1. Text Collection (Corpus Creation)

The first step in text mining is the collection of textual data from various
sources such as documents, emails, web pages, social media posts, and
customer feedback. The complete set of collected documents is known as
a corpus.
Example: Customer reviews collected from an e-commerce website like
product ratings and comments.

2. Text Preprocessing (Cleaning)

Text preprocessing is a crucial step that improves data quality and
reduces complexity. Since raw text contains noise and irrelevant
information, it must be cleaned before analysis.
This step includes:
• Tokenization – splitting text into individual words or terms
• Stop-word removal – removing common words like is, the, and
• Stemming / Lemmatization – reducing words to their root form (e.g.,
running → run)
• Noise removal – removing symbols, numbers, punctuation, and HTML
tags
 The main purpose of preprocessing is to improve accuracy and efficiency
of text analysis.

3. Text Transformation / Feature Extraction

In this step, cleaned text is converted into a numerical form so that
mathematical and statistical techniques can be applied.
A commonly used method is the Term–Document Matrix (TDM),
where:
• Rows represent terms
• Columns represent documents
• Values represent term frequency (TF) or TF-IDF
This transformation enables machine learning and data mining algorithms
to work on text data.

4. Text Mining Techniques (Analysis)

After feature extraction, various text mining techniques are applied to
discover patterns and insights from text.
Common techniques include:
• Text Classification – assigning documents to predefined categories
• Clustering – grouping similar documents
• Association Rule Mining – finding relationships among words or terms
• Sentiment Analysis – identifying opinions such as positive, negative, or
neutral
• Topic Modeling – discovering hidden topics within documents

5. Pattern Evaluation and Interpretation

In this step, the extracted patterns are evaluated to check their relevance
and usefulness. Irrelevant or redundant results are removed, and
meaningful patterns are validated using domain knowledge.
Example: Identifying major customer complaints such as delivery delay
or poor quality from product reviews.

6. Knowledge Presentation
The final step is presenting the extracted knowledge in an easy-to-
understand format. Results are presented using charts, dashboards,
reports, and word clouds. This helps managers and decision-makers take
appropriate actions based on insights.
3. Explain Natural Language Processing

Natural Language Processing (NLP) is a branch of Artificial

Intelligence that enables computers to process, understand, analyze,
and generate human language in the form of text or speech. NLP helps
machines extract information, identify patterns, and respond intelligently
to human language.

Objectives of NLP
The main objectives of Natural Language Processing are:
• To enable computers to understand and interpret human language
• To convert unstructured text into structured and meaningful information
• To support text analytics and text mining applications
• To improve human–computer interaction through natural communication

Components of NLP
1. Lexical Analysis
Lexical analysis breaks text into smaller units called tokens and identifies
the structure of words.
Example:
“Data analytics” → Data, analytics

2. Syntactic Analysis (Parsing)

Syntactic analysis examines the grammatical structure of sentences
using grammar rules. It checks whether a sentence is syntactically correct.
 Example:
Subject–Verb–Object relationship in a sentence.

3. Semantic Analysis
Semantic analysis focuses on the meaning of words and sentences. It
helps resolve ambiguity in language.
Example:
The word “bank” may refer to a financial bank or a river bank depending
on context.

4. Discourse Integration
Discourse integration considers the context of previous sentences to
understand the meaning of the current sentence. It helps maintain
continuity across multiple sentences.

5. Pragmatic Analysis
Pragmatic analysis interprets language based on real-world context and
intent. It determines what the speaker actually means rather than what is
explicitly stated.

NLP Techniques
Common techniques used in NLP include:
• Tokenization
• Stop-word removal
• Stemming and Lemmatization
• Part-of-Speech (POS) tagging
• Named Entity Recognition (NER)
• Parsing

Applications of NLP
NLP is widely used in many real-world applications such as:
• Chatbots and virtual assistants
• Machine translation (e.g., Google Translate)
• Sentiment analysis of reviews and social media
• Text summarization
• Speech recognition systems
• Spam filtering
• Information retrieval and search engines
4. Explain applications of Text Mining

Text Mining is widely used to extract meaningful information and hidden

patterns from large volumes of unstructured text data such as emails,
documents, reviews, news articles, and social media content. It helps
organizations convert textual data into actionable knowledge for better
decision-making.

Major Applications of Text Mining

1. Marketing and Customer Relationship Management

• Analyzes customer reviews, feedback, and surveys
• Identifies customer preferences and expectations
• Supports sentiment analysis for brand monitoring
• Improves targeted marketing and customer satisfaction
Example: Analyzing online product reviews to improve product quality.

2. Social Media Analytics

• Monitors public opinion on social media platforms
• Detects emerging trends and viral topics
• Identifies influencers and customer sentiments
Example: Analyzing Twitter data to understand public reaction to a
product launch.

3. Healthcare and Biomedical Research

• Extracts information from medical records and clinical notes
• Supports disease diagnosis and treatment analysis
• Helps in drug discovery and medical research
Example: Mining research papers to identify disease–drug relationships.

4. Fraud Detection and Risk Management

• Identifies suspicious patterns in insurance claims and financial documents
• Detects fraud and compliance violations
• Supports credit risk analysis
Example: Detecting fraudulent insurance claims using claim
descriptions.

5. Legal and E-Discovery

• Analyzes legal documents, contracts, and case files
• Supports legal research and compliance
• Helps in document classification and retrieval
Example: Identifying relevant legal documents during investigations.

6. Spam Detection and Email Filtering

• Classifies emails as spam or legitimate
• Improves cybersecurity and productivity
• Uses classification and pattern recognition
Example: Gmail spam filtering system.

7. Business Intelligence and Decision Support

• Converts unstructured text into structured insights
• Supports strategic and operational decisions
• Enhances reporting and analytics
Example: Mining internal company reports for performance analysis.

8. Academic and Research Analysis

• Analyzes large volumes of research papers
• Identifies trends, citations, and research gaps
• Supports literature surveys
Example: Text mining for systematic literature review.

9. Web Mining and Search Engines

• Improves search result relevance
• Analyzes web content and user queries
• Supports recommendation systems
Example: Google search engine ranking and suggestions.

10. Customer Support and Chatbots

• Analyzes customer queries and complaints
• Improves automated response systems
• Enhances service quality
Example: AI-powered customer service chatbots.

5. Explain Sentiment Analysis and its types

Sentiment Analysis is the process of analyzing text data to determine the

sentiment or opinion expressed by the author. It identifies whether the
sentiment is positive, negative, or neutral, and in some cases measures
the strength or intensity of the sentiment.
 Sentiment Analysis is also known as Opinion Mining, Subjectivity
Analysis, and Appraisal Extraction.

Need for Sentiment Analysis

With the rapid growth of the internet and social media, a huge volume of
opinion-rich data such as reviews, comments, tweets, and feedback is
generated every day. Traditional surveys are slow, costly, and limited in
scope. Sentiment analysis helps organizations monitor opinions in real
time, understand customer perceptions, and support customer-centric
decision making.

Types of Sentiment Analysis

1. Based on Sentiment Polarity
This is the most common classification.
• Positive Sentiment – expresses satisfaction or approval
Example: “This product is excellent.”
• Negative Sentiment – expresses dissatisfaction or criticism
Example: “The service is very poor.”
• Neutral Sentiment – expresses no strong opinion
Example: “The product was delivered yesterday.”

2. Explicit vs Implicit Sentiment

Explicit Sentiment
The opinion is directly stated.
Example: “The phone camera is amazing.”
Implicit Sentiment
The opinion is implied through facts or situations.
Example: “The battery lasts only two hours.”
(This implies a negative sentiment.)

3. Based on Level of Analysis

a) Word-Level Sentiment Analysis
Determines the sentiment of individual words using sentiment lexicons or
dictionaries.
b) Sentence-Level Sentiment Analysis
Determines the sentiment of an entire sentence and handles negations and
modifiers.
Example: “The phone is not good.”
 c) Document-Level Sentiment Analysis
Determines the overall sentiment of a complete document such as a
review or article.

4. Based on Learning Approach

a) Lexicon-Based (Unsupervised) Sentiment Analysis
Uses predefined sentiment dictionaries such as WordNet or
SentiWordNet.
No training data is required.
Advantages:
• Easy to implement
• Domain-independent
Disadvantages:
• Lower accuracy for complex or context-dependent language
b) Machine Learning–Based (Supervised) Sentiment Analysis
Uses labeled training data to build models.
Algorithms used:
Naïve Bayes, Support Vector Machine (SVM), Decision Trees, Neural
Networks
Advantages:
• Higher accuracy
• Can learn complex patterns
Disadvantages:
• Requires large labeled datasets

Applications of Sentiment Analysis

Sentiment analysis is widely used in:
• Product and brand analysis
• Customer feedback analysis
• Social media monitoring
• Financial market prediction
• Political opinion and election analysis

6. Explain Web Mining and its types

Web Mining is the process of extracting and discovering meaningful

patterns and information from Web data, including web content, web
structure, and web usage data.
 Web mining goes beyond simple keyword search by identifying hidden
relationships, trends, and user behaviors on the Web.

1. Web Content Mining

Web Content Mining focuses on extracting useful information from the
content of web pages. It deals with the data that users see on the web.
Data Sources
• Text data (HTML, XML)
• Images, audio, and video
• Metadata and tags
Techniques Used
• Text mining
• Natural Language Processing (NLP)
• Information extraction
Applications
• News and opinion mining
• Sentiment analysis
• Competitive intelligence
• Product review analysis
Example:
Extracting customer opinions and ratings from online product reviews on
e-commerce websites.

2. Web Structure Mining

Web Structure Mining analyzes the link structure (hyperlinks) of the
web to understand relationships between web pages.
 Key Concepts
• Authority pages – Highly trusted and frequently referenced pages
• Hub pages – Pages that link to many authoritative pages
Techniques Used
• Link analysis
• Page ranking algorithms such as PageRank and HITS
Applications
• Search engine ranking
• Identifying important and relevant web pages
• Understanding web communities
Example:
Search engines ranking websites based on the number and quality of
backlinks.

3. Web Usage Mining

Web Usage Mining, also known as Web Analytics, analyzes user
behavior data generated when users interact with websites.
Data Sources
• Server log files
• Clickstream data
• Cookies and session data
Techniques Used
• Pattern discovery
• Session analysis
• User profiling
Applications
• Website personalization
• Recommendation systems
• Traffic analysis
• Improving website usability
Example:
Analyzing customer navigation paths on an e-commerce website to
recommend products.

7. Explain Web Usage Mining.

Web Usage Mining is the process of extracting and analyzing user

interaction data (clickstream data) collected from web servers, proxy
servers, and client browsers to understand, model, and predict user
 behavior on the Web.
It is also commonly referred to as Web Analytics.

Sources of Web Usage Data

Web usage data is collected from multiple sources:
1. Web Server Logs
Contain records of page requests such as timestamps, IP addresses, URLs
accessed, and request methods.
2. Proxy Server Logs
Capture requests from multiple users within an organization and help
analyze group-level behavior.
3. Client-Side Data
Collected using cookies, browser history, session data, and JavaScript
page tagging to track individual user behavior accurately.

Web Usage Mining Process

1. Data Collection
Raw clickstream data is collected from server logs and tracking tools.
Includes page views, session duration, navigation paths, and user actions.

2. Data Preprocessing
This step cleans and prepares data for analysis. It includes:
• Data cleaning (removing noise, bots, and irrelevant requests)
• User identification
• Session identification
• Path completion (handling missing page references)

3. Pattern Discovery
Data mining techniques are applied to discover meaningful patterns:
• Association rule mining – pages frequently visited together
• Clustering – grouping similar users
• Classification – predicting user behavior
• Sequential pattern mining – identifying navigation sequences

4. Pattern Analysis
Discovered patterns are filtered, interpreted, and visualized using
dashboards and reports. This step converts patterns into actionable
insights for decision-making.

Applications of Web Usage Mining

 Web usage mining is widely used for:
• Website personalization
• Recommendation systems
• Improving website usability
• Targeted advertising
• Customer behavior analysis
• Traffic and conversion analysis
Example:
Amazon recommending products based on a user’s browsing and
purchase history.

Advantages of Web Usage Mining

• Improves customer satisfaction
• Enhances website performance
• Supports data-driven marketing
• Increases conversion rates
• Helps retain customers

Limitations of Web Usage Mining

• Privacy and data protection concerns
• Incomplete or noisy data
• Difficulty in accurate user identification
• Handling very large data volumes

8. “Explain the working of search engines.”

A search engine is a software system designed to search, retrieve, and

rank relevant web pages or documents from the World Wide Web in
response to a user query. Examples include Google, Bing, and Yahoo. The
main goal of a search engine is to provide accurate and relevant results
quickly.
 1.

Development Cycle
This cycle focuses on collecting, processing, and indexing web pages
before users submit queries.
a) Web Crawler (Spider)
A web crawler automatically browses the World Wide Web to discover
and download web pages.
• Starts with a set of seed URLs
• Fetches web pages and follows hyperlinks to find new pages
• Stores fetched pages for further processing
Example: Googlebot

b) Document Preprocessing
Downloaded web pages are converted into a standard, clean format.
• Separates text, metadata, and hyperlinks
• Removes HTML tags, scripts, and noise
• Normalizes content for analysis

c) Document Indexer
The indexer applies text mining and NLP techniques to make
documents searchable.
• Tokenization
• Stop-word removal
• Stemming
• Builds an Inverted Index (or Term–Document Matrix)
• Stores indexed data in a searchable database

2. Response Cycle
This cycle is activated when a user submits a query.
a) Query Analyzer
Processes the user query using steps similar to document indexing:
• Tokenization
• Stop-word removal
• Stemming
• Query normalization
The query is converted into a machine-understandable form.

b) Document Matcher
Matches the processed query against the indexed documents.
• Identifies documents that contain relevant terms
• Retrieves candidate pages from the index

c) Ranking Module
Ranks retrieved documents based on relevance using multiple factors:
• Keyword relevance (TF-IDF)
• Link analysis (PageRank, authority/hub concepts)
• User behavior signals (click-through rate, freshness)
Higher-ranked pages are considered more relevant.

d) Results Presentation
Displays ranked results to the user with:
• Page title
• URL
• Snippet (summary of content)

Evaluation of Search Engines

Search engines are evaluated based on:
• Effectiveness – relevance and quality of results
• Efficiency – speed of response

Advantages of Search Engines

• Quick access to vast information
• Supports decision-making and research
• Enables e-commerce and online services
• Enhances learning and knowledge discovery
9. Explain Social Media Analytics and its applications

Social Media Analytics is the process of collecting, analyzing, and

interpreting data generated on social media platforms (such as posts,
comments, likes, shares, tweets, and reviews) to understand user
behavior, opinions, trends, and relationships. It helps organizations gain
insights for marketing, customer engagement, brand management,
and decision-making.
Social media analytics uses techniques from text analytics, sentiment
analysis, network analysis, and predictive analytics to extract
actionable insights from unstructured social data.

Key Objectives of Social Media Analytics

• Understand public opinion and customer sentiment
• Monitor brand reputation and online conversations
• Identify trends, influencers, and communities
• Support marketing and business decisions
• Enable real-time response to events and feedback

Process of Social Media Analytics (Brief)

1. Data Collection – Gather data from platforms like posts, tweets,
comments, hashtags
2. Data Preprocessing – Clean text, remove noise, handle duplicates
3. Analysis – Apply sentiment analysis, topic modeling, network analysis
4. Visualization & Reporting – Dashboards, charts, alerts
5. Decision Making – Strategy formulation and action

Applications of Social Media Analytics

1. Brand Monitoring and Reputation Management
Organizations track mentions, reviews, and sentiments to understand how
the brand is perceived online.
Example: Detecting negative reviews early to prevent brand damage.

2. Sentiment Analysis
Analyzes customer opinions as positive, negative, or neutral to measure
satisfaction.
Example: Analyzing tweets about a product launch.
 3. Marketing and Campaign Analysis
Evaluates the effectiveness of marketing campaigns by measuring
engagement, reach, and conversion.
Example: Measuring likes, shares, and comments for an ad campaign.

4. Customer Feedback and Service Improvement

Identifies common complaints, suggestions, and expectations from
customers.
Example: Improving delivery service based on repeated complaints.

5. Trend Detection and Market Research

Detects emerging topics, trends, and consumer interests in real time.
Example: Identifying trending products or hashtags.

6. Influencer Identification
Finds influential users who can impact opinions and purchasing
decisions.
Example: Selecting influencers for product promotions.

7. Political and Public Opinion Analysis

Analyzes public sentiment toward policies, leaders, or events.
Example: Election sentiment analysis using social media posts.

8. Risk and Crisis Management

Helps detect potential crises early by monitoring sudden spikes in
negative sentiment.
Example: Early detection of backlash against a company decision.

Benefits of Social Media Analytics

• Real-time insights
• Better customer engagement
• Data-driven marketing strategies
• Improved brand reputation management