‭UNIT-4‬

‭Search Statements & Binding‬

‭1.‬‭Search Statements:‬
‭○‬ ‭What are they?‬‭These are the instructions given‬
‭by users to find specific information within items‬
‭(documents, articles, etc.).‬
‭○‬ ‭How are they created?‬‭Users write these using‬
‭either Boolean logic (AND, OR, NOT) or in natural‬
‭language (like regular sentences).‬
‭○‬ ‭Example:‬‭A user might write "renewable energy‬
‭AND solar power" to find items containing both‬
‭terms.‬
‭2.‬‭Weighting Concepts:‬
‭○‬ ‭What does it mean?‬‭Users can indicate which‬
‭concepts in their search are more important.‬
‭○‬ ‭Example:‬‭In the search "renewable energy AND‬
‭solar power," the user might indicate that "solar‬
‭power" is more important by giving it a higher‬
‭weight.‬
‭3.‬‭Binding to Vocabulary & Experience:‬
‭○‬ ‭What is it?‬‭The search statement reflects the‬
‭user's knowledge and vocabulary.‬
‭○‬ ‭Example:‬‭A scientist might use technical terms,‬
‭while a student might use simpler language.‬
‭4.‬‭Subset of Items:‬
‭○‬ ‭What is it?‬‭The search statement helps narrow‬
‭down the total number of items to those relevant to‬
‭the user's needs.‬
 ‭○‬ ‭Example:‬‭From thousands of articles, the search‬
‭"renewable energy AND solar power" might‬
‭narrow it down to a few hundred relevant ones.‬
‭5.‬‭System Parsing:‬
‭○‬ ‭What happens?‬‭The search system translates‬
‭the user's query into its own language (meta‬
‭language) to process it.‬
‭○‬ ‭Example:‬‭The system might convert "renewable‬
‭energy AND solar power" into specific commands‬
‭it understands.‬
‭Types of Systems‬

‭1.‬‭Statistical Systems:‬
‭○‬ ‭How do they work?‬‭They analyze the frequency‬
‭of words in the search statement.‬
‭○‬ ‭Example:‬‭If "solar power" appears frequently in‬
‭the database, it gets more weight in the search‬
‭results.‬
‭2.‬‭Natural Language Systems:‬
‭○‬ ‭How do they work?‬‭They understand the‬
‭grammar and meaning of the search statement.‬
‭○‬ ‭Example:‬‭They can distinguish between "bank" as‬
‭in a financial institution and "bank" as in a‬
‭riverbank.‬
‭3.‬‭Concept Systems:‬
‭○‬ ‭How do they work?‬‭They map the search terms‬
‭to broader concepts used for indexing items.‬
‭○‬ ‭Example:‬‭The system might link "solar power" to‬
‭broader concepts like "renewable energy" or‬
‭"sustainable technology."‬
‭Statistics in Weighting‬

‭1.‬‭Document Frequency:‬
‭○‬ ‭What is it?‬‭How often a term appears in the‬
‭database.‬
‭○‬ ‭Example:‬‭If "renewable" appears in many‬
‭documents, it might be given less weight compared‬
‭to a less common term like "geothermal."‬
‭2.‬‭Total Frequency:‬
‭○‬ ‭What is it?‬‭The total number of times a term‬
‭appears.‬
‭○‬ ‭Example:‬‭"Energy" might have a high total‬
‭f requency because it appears in many contexts.‬
‭Using Parentheses‬

‭1.‬‭What do they do?‬‭Indicate expansions using a‬

‭thesaurus.‬
‭○‬ ‭Example:‬‭Searching for "renewable energy" might‬
‭also include terms like "green energy" or‬
‭"sustainable power" using a thesaurus.‬
‭Length of Search Statements‬

‭1.‬‭Impact on Results:‬
‭○‬ ‭Short Queries:‬‭Harder for the system to find‬
‭relevant items because they are too broad.‬
‭○‬ ‭Long Queries:‬‭Easier for the system to find‬
‭relevant items because they are more specific.‬
‭○‬ ‭Example:‬‭"Energy" vs. "renewable energy‬
‭sources in Europe" - the longer query is more‬
‭specific and helps find better matches.‬
I‭n summary, creating effective search statements involves‬
‭using the right language, weighting important terms, and‬
‭understanding how different systems process these‬
‭searches to find the most relevant information. Longer and‬
‭more detailed queries generally yield better results.‬
 ‭Searching the Internet & Hypertext‬‭Explained‬

‭Internet Searching Mechanisms:‬

‭●‬ ‭Techniques:‬‭Servers on the Internet create indexes‬

‭of items and allow searches. Popular nodes include‬
‭Yahoo, AltaVista, and Lycos.‬
‭●‬ ‭Example:‬‭These systems visit many websites, retrieve‬
‭text data, and index it for search purposes.‬

‭Design Decisions:‬

‭●‬ ‭Data Retrieval:‬‭Different levels of data retrieval‬‭and‬

‭user access philosophies are used.‬
‭●‬ ‭Example:‬‭Lycos retrieves home pages for indexing,‬
‭while AltaVista indexes all text at a site.‬

‭Index Creation:‬

‭●‬ ‭Process:‬‭Retrieved text is used to create an index,‬

‭storing URLs for user access.‬
‭●‬ ‭Example:‬‭If you search "solar energy," the system‬
‭provides a list of URLs where the term appears.‬

‭Ranking Algorithms:‬

‭●‬ ‭Ranking:‬‭Simple algorithms based on word‬

‭occurrence help display relevant results.‬
‭●‬ ‭Example:‬‭More frequently mentioned terms in the text‬
‭are ranked higher.‬

‭Intelligent Agents:‬
 ‭●‬ ‭Function:‬‭These agents search the Internet‬
‭independently, locating information based on‬
‭user-defined needs.‬
‭●‬ ‭Example:‬‭An agent might search for the latest news‬‭on‬
‭"climate change" while you are offline.‬

‭Key Characteristics of Intelligent Agents:‬

‭1.‬‭Autonomy:‬
‭○‬ ‭Definition:‬‭Operate without human interaction,‬
‭control internal states, and make independent‬
‭decisions.‬
‭○‬ ‭Example:‬‭An agent autonomously searches for‬
‭"renewable energy" sites.‬
‭2.‬‭Communication Ability:‬
‭○‬ ‭Definition:‬‭Must communicate with information‬
‭sites using a universal language.‬
‭○‬ ‭Example:‬‭The agent uses HTTP protocols to‬
‭f etch web pages.‬
‭3.‬‭Capacity for Cooperation:‬
‭○‬ ‭Definition:‬‭Agents cooperate to perform tasks.‬
‭○‬ ‭Example:‬‭Multiple agents share findings on "solar‬
‭panel technology."‬
‭4.‬‭Capacity for Reasoning:‬
‭○‬ ‭Scenarios:‬
‭■‬ ‭Rule-Based:‬‭Uses predefined conditions‬
‭and actions.‬
‭■‬ ‭Example:‬‭An agent searches for "electric‬
‭cars" based on set rules.‬
‭■‬ ‭Knowledge-Based:‬‭Uses past conditions‬
‭and actions to deduce future actions.‬
 ‭■‬ ‭Example:‬‭An agent remembers previous‬
‭searches and improves future ones.‬
‭■‬ ‭Artificial Evolution:‬‭Creates new agents‬
‭with enhanced logic.‬
‭■‬ ‭Example:‬‭An agent evolves to better search‬
‭f or "AI research papers."‬
‭5.‬‭Adaptive Behavior:‬
‭○‬ ‭Definition:‬‭Assesses current state and decides‬
‭on actions.‬
‭○‬ ‭Example:‬‭An agent changes its search strategy‬
‭based on search success.‬
‭6.‬‭Trustworthiness:‬
‭○‬ ‭Definition:‬‭Users must trust the agent to act in‬
‭their interest.‬
‭○‬ ‭Example:‬‭An agent reliably finds and delivers‬
‭relevant research articles.‬

I‭n summary, internet searching involves sophisticated‬

‭techniques and intelligent agents that autonomously retrieve,‬
‭index, and rank information based on user-defined needs.‬
‭These agents possess characteristics like autonomy,‬
‭communication ability, cooperation, reasoning, adaptive‬
‭behaviour, and trustworthiness to efficiently locate relevant‬
‭information.‬
‭Closure‬

‭ rinciple:‬‭W hen we see an incomplete shape, our brain‬

P
‭f ills in the missing parts to see it as a whole.‬

‭ xample:‬‭If you see a drawing of a circle made with dashed‬

E
‭lines, your brain still sees it as a full circle, even though‬
‭parts of the line are missing.‬
‭Tree Structure‬

‭ xplanation:‬‭A hierarchical structure that organizes‬‭data in‬

E
‭a parent-child relationship. Each element (node) has one‬
‭parent and can have multiple children.‬

‭Example:‬

‭●‬ ‭File System:‬‭Folders and files on a computer are‬

‭organized in a tree structure, where each folder can‬
‭contain subfolders and files.‬
‭●‬ ‭Organizational Chart:‬‭Shows the hierarchy within a‬
‭company, from the CEO down to individual employees.‬
‭Cone-Tree‬

‭ xplanation:‬‭A 3D visualization technique used to‬‭display‬

E
‭hierarchical data. Nodes are arranged in a cone shape,‬
‭making it easier to explore large trees by rotating the cone.‬
‭Example:‬

‭●‬ ‭Large Organizational Structures:‬‭Helps visualize‬

‭and navigate complex organizational charts in a 3D‬
‭space.‬
‭●‬ ‭Website Navigation:‬‭Can be used to represent‬
‭website structures, allowing users to see the hierarchy‬
‭of web pages.‬
‭Perspective Wall‬

‭ xplanation:‬‭A visualization technique that combines a‬

E
‭f ocus region with a distorted surrounding context, like a‬
‭f isheye lens. It allows users to focus on one part of the‬
‭information while still seeing the broader context.‬

‭Example:‬

‭●‬ ‭T imelines:‬‭Viewing detailed events in the center while‬

‭keeping the entire timeline visible in the background.‬
‭●‬ ‭Large Documents:‬‭Focus on a particular section of a‬
‭document while having a general view of the entire‬
‭document.‬
‭Tree Maps‬

‭ xplanation:‬‭A method of displaying hierarchical data‬

E
‭using nested rectangles. Each rectangle represents a node,‬
‭and its size is proportional to a specific data dimension.‬

‭Example:‬
‭●‬ ‭Disk Space Usage:‬‭Visualizes how much space each‬
‭f older and file takes up on a hard drive.‬
‭●‬ ‭Stock Market Data:‬‭Shows the market capitalization‬
‭of companies, where each rectangle represents a‬
‭company and its size represents its market value.‬
‭Envision System‬

‭ xplanation:‬‭A visualization system designed to help‬‭users‬

E
‭understand and analyze large datasets. It typically includes‬
‭various tools for interactive exploration and analysis.‬

‭Example:‬

‭●‬ ‭Scientific Data Analysis:‬‭Helps researchers‬

‭visualize complex datasets to identify patterns and‬
‭trends.‬
‭●‬ ‭Business Intelligence:‬‭Allows users to explore and‬
‭analyze sales data, customer behavior, and other‬
‭business metrics.‬
‭Document Content Analysis & Retrieval System (DCARS)‬
‭ xplanation:‬‭A system designed to analyze and retrieve‬
E
‭documents based on their content. It typically uses text‬
‭analysis and search algorithms to help users find relevant‬
‭documents quickly.‬

‭Example:‬

‭●‬ ‭Legal Document Retrieval:‬‭Helps lawyers find‬

‭relevant case files and legal documents based on‬
‭keywords and content analysis.‬
‭●‬ ‭Academic Research:‬‭Assists researchers in finding‬
‭relevant papers and articles based on content analysis‬
‭of academic databases.‬
‭City Space‬

‭ xplanation:‬‭A virtual reality or 3D visualization‬‭system‬

E
‭that represents a city or urban environment. It can be used‬
‭f or planning, navigation, and analysis of urban data.‬

‭Example:‬

‭●‬ ‭Urban Planning:‬‭Helps city planners visualize and‬

‭simulate urban development projects.‬
‭●‬ ‭Tourism:‬‭Provides virtual tours of cities, allowing‬‭users‬
‭to explore landmarks and attractions in a virtual‬
‭environment.‬