In a search engine, the index stores:Raw documentsDocument metadata and keywordsMultimedia filesUser profilesB) Document metadata and keywordsIndexes map documents to keywords for faster retrieval.
PageRank algorithm was developed by:YahooGoogleMicrosoftIBMB) GooglePageRank measures importance of web pages based on links.
Precision in information retrieval measures:Relevant retrieved / Total retrievedRelevant retrieved / Total relevantTotal retrieved / Relevant retrievedNone of the aboveA) Relevant retrieved / Total retrievedPrecision indicates the fraction of retrieved results that are relevant.
Recall in information retrieval measures:Relevant retrieved / Total relevantRelevant retrieved / Total retrievedPrecision + relevanceTotal documents / RetrievedA) Relevant retrieved / Total relevantRecall measures completeness of retrieved relevant documents.
Which model represents documents and queries as vectors in multidimensional space?Boolean ModelVector Space ModelProbabilistic ModelLanguage ModelB) Vector Space ModelVector Space Model uses term vectors and cosine similarity.
In IR, precision measures:Relevant retrieved / Total retrievedRelevant retrieved / Total relevantTotal retrieved / Total documentsTotal relevant / Total documentsA) Relevant retrieved / Total retrievedPrecision reflects how accurate search results are.
A stop word in IR refers to:A frequently used common wordA keyword for indexingAn encrypted termA synonymA) A frequently used common wordWords like “the,” “and,” “is” are ignored to reduce index size.
Inverted index stores:Documents in sequenceMapping of terms to documentsFile structurePage linksB) Mapping of terms to documentsInverted indexes quickly retrieve documents containing specific words.
TF-IDF is used to:Rank term importanceCompress filesMeasure redundancyClassify documentsA) Rank term importanceTF-IDF balances term frequency with rarity across documents for better search relevance.
TF-IDF stands for:Term Frequency – Inverse Document FrequencyTransfer Function – Inverted Data FlowTerm Field – Internal Data FactorTotal File – Indexing Data FormulaA) Term Frequency – Inverse Document FrequencyTF-IDF measures importance of terms in documents.
Precision in IR measures:Relevant retrieved / total retrievedRelevant retrieved / total relevantTotal retrieved / all documentsSpeed of searchA) Relevant retrieved / total retrievedPrecision measures correctness of retrieved results.
Recall measures:Fraction of relevant documents retrievedFraction of retrieved documents relevantIndexing timeQuery lengthA) Fraction of relevant documents retrievedRecall emphasizes completeness.
PageRank was developed for:Database optimizationWeb search rankingImage recognitionBlockchainB) Web search rankingPageRank by Google ranks web pages using link analysis.
In IR, stemming reduces words to:Root/base formUppercase formRandom tokensCompound wordsA) Root/base formExample: "running" → "run".
Recall is high when:Most relevant docs are retrievedFew irrelevant docs are retrievedSpeed of query is maximumIndex is compactA) Most relevant docs are retrievedRecall measures completeness.
Cosine similarity is widely used for:Document similarityFile compression) Data encryptionCloud storageA) Document similarityCosine similarity compares document vectors.
In IR, query expansion means:Adding related terms to improve retrievalDeleting terms from queryCompressing the indexReducing document sizeA) Adding related terms to improve retrievalIt increases recall by broadening search terms.
Which is an evaluation metric in IR?PrecisionLatencyThroughputBandwidthA) PrecisionPrecision measures proportion of relevant retrieved docs.
TF-IDF helps by:Measuring word importance in documentsEncrypting textSorting web pages alphabeticallyCounting stop wordsA) Measuring word importance in documentsTF-IDF balances term frequency with inverse document frequency.
Precision in IR measures:Relevant results retrieved / Total retrievedTotal retrieved / Total in databaseTime taken to searchUser satisfactionA) Relevant results retrieved / Total retrievedPrecision = fraction of correct results returned.
A recall-oriented system focuses on:Retrieving fewer resultsRetrieving all relevant results, even with noiseFaster searches onlyRemoving irrelevant documents completelyB) Retrieving all relevant results, even with noiseRecall emphasizes completeness over precision.
An inverted index stores:Keywords mapped to document IDsDocument IDs mapped to usersUser logsPage load timesA) Keywords mapped to document IDsCore data structure for search engines.
A search engine index stores:Raw HTMLProcessed terms and document linksServer logsImages B) Processed terms and document linksIndexes map terms to documents for faster searching.
Stemming is used to:Shorten words to base formTranslate queriesRank documentsCompress dataA) Shorten words to base formIt helps group word variants like “connect,” “connected,” “connection.”
A search engine “crawler” does:Query rankingWeb page indexingCache cleaning EncryptionB) Web page indexingCrawlers discover and store webpage data.
The Boolean model uses:AND/OR/NOT operationsProbabilitiesNeural embeddingsRandom searchA) AND/OR/NOT operationsClassical model for keyword matching.
Precision measures:Relevant results retrieved / Total results retrievedSpeed Recall accuracySearch costA) Relevant results retrieved / Total results retrievedHigh precision means fewer false positives.
In Information Retrieval, which technique is primarily used to measure how relevant a document is to a user’s query by considering both term frequency and document rarity?Latent Semantic Indexing (LSI)TF-IDFWord EmbeddingCosine SimilarityB) TF-IDFTF-IDF weighs a term’s importance by considering how often it appears in a document (TF) versus how rare it is across all documents (IDF). This helps rank search results based on relevance — a key concept in modern information retrieval systems.
The inverted index is a data structure that:Maps terms to documentsMaps users to search history: Encrypts textStores binary treesA) Maps terms to documentsIt speeds up keyword searches by linking words to document IDs.
The PageRank algorithm is replaced in modern search engines by:Neural ranking modelsManual indexingStatic rankingURL countingA) Neural ranking modelsMachine learning–based rankers now evaluate user intent and content semantics.
What is precision in IR systems?Relevant documents retrieved / Total retrieved documents Total relevant documents / Total documentsFalse positives / True positivesRecall valueA) Relevant documents retrieved / Total retrieved documentsPrecision measures retrieval accuracy of relevant results.
What is the main goal of a ranking algorithm?Sorting documents randomlyPrioritizing relevant resultsCompressing dataDeleting duplicatesB) Prioritizing relevant resultsRanking orders documents by predicted relevance to a query.
What is stemming in information retrieval?Indexing imagesReducing words to root formCounting sentencesTranslating textB) Reducing words to root formStemming helps improve search matches by generalizing word forms
What is an inverted index used for?Reverse engineeringFast document search Database backupCompressionB) Fast document searchInverted index maps terms to document IDs for efficient retrieval.
Which of the following best describes Information Retrieval (IR)?Extracting information from structured databasesFinding relevant information from unstructured dataEncrypting digital contenCompiling code for executionB) Finding relevant information from unstructured dataIR systems focus on searching and retrieving meaningful documents or data from unstructured sources like web pages and text.
The main component of an IR system that stores words and their document references is called:Hash tableInverted indexTrieDatabase schemaB) Inverted indexAn inverted index maps terms to their occurrences in documents, making search fast and efficient.
The process of reducing words like “running” and “runs” to “run” is known as:ParsingTokenizationStemmingClusteringC) StemmingStemming normalizes word forms to their root, improving recall in search results.
In a search engine, ranking algorithms mainly aim to:Compress documentsPrioritize the most relevant resultsEliminate duplicate queriesIncrease database sizeB) Prioritize the most relevant resultsRanking ensures that the most relevant documents appear first according to query relevance metrics.
The vector space model represents documents as:ImagesVectors of termsTablesTreesB) Vectors of termsEach document is a vector where dimensions correspond to words.
Precision in IR measures:Fraction of relevant documents retrievedTotal documents retrievedProcessing timeQuery speedA) Fraction of relevant documents retrievedPrecision focuses on result accuracy.
The main goal of an information retrieval system is to:Store web pagesRetrieve relevant documentsCompress data Index images onlyB) Retrieve relevant documentsIR systems aim to fetch documents matching user queries.
A search engine’s “indexer” performs:RankingQuery expansion Document parsing and storageUser profilingC) Document parsing and storageIndexing prepares searchable representations of data.
The PageRank algorithm’s modern successor uses:Content embeddingsKeyword matching Click-through rates onlyBoolean filtersA) Content embeddingsModern IR uses vector-based ranking instead of link-only methods.
What is the main objective of an information retrieval system?Data encryptionData retrievalData visualizationData cleaningB) Data retrievalThe system’s primary goal is to locate relevant information from large datasets efficiently.
What is “query expansion” used for?Adding synonyms to improve search resultsReducing index sizeRemoving stop wordsSorting search resultsA) Adding synonyms to improve search resultsQuery expansion enhances recall by adding related terms to the user’s query.
TF-IDF weighting helps in:Compressing documentsRanking relevance of termsCreating hyperlinksMeasuring document sizeB) Ranking relevance of termsTF-IDF assigns higher importance to terms that are frequent in a document but rare across the corpus.
What is the main advantage of inverted index structures?Faster document updatesEfficient search term lookupLower memory useAutomatic summarizationB) Efficient search term lookupInverted indexes map words to document lists, speeding up retrieval.
Precision in information retrieval refers to:Number of relevant documents retrievedTotal number of retrieved documentsRatio of relevant retrieved to total retrievedRecall scoreC) Ratio of relevant retrieved to total retrievedPrecision measures accuracy of search results by comparing relevant results to total output.
The vector space model represents:Words as nodesDocuments as vectorsQueries as keywordsData as treesB) Documents as vectorsEach document is represented as a vector in term space for similarity comparison.
Cosine similarity is used to:Compare two vectorsNormalize TF-IDF valuesSort rankingRemove stop wordsA) Compare two vectorsCosine similarity calculates the angle between document and query vectors to determine relevance.
What does TF–IDF measure?Query length Term importanceFile sizeSystem delayB) Term importanceTF–IDF highlights how relevant a word is within a document set.
Inverted index stores:Documents in orderTerms and their document occurrencesFile metadataPage ranksB) Terms and their document occurrencesIt maps words to documents that contain them for fast searching.
What is precision in information retrieval?Correct results divided by total results retrievedTotal documents retrievedFalse negativesRecall inverseA) Correct results divided by total results retrievedPrecision measures search accuracy.
The main goal of a search engine is to:Store dataRank and retrieve relevant informationMonitor usersCompress filesB) Rank and retrieve relevant informationSearch engines use algorithms to return the most relevant documents.
Which algorithm measures text similarity?TF–IDF K-MeansCosine similarityAprioriC) Cosine similarityIt calculates similarity between document vectors using their angle.
Inverted index stores:User IDsWord-to-document mappings MetadataURL rankingsB) Word-to-document mappingsIt enables fast retrieval of documents containing specific terms.
TF-IDF helps to:Compress dataRank document importanceRemove stop wordsCluster dataB) Rank document importanceTF-IDF weighs terms that are frequent in a document but rare across documents.
Recall measures:Accuracy Fraction of relevant items retrievedPrecision onlyProcessing timeB) Fraction of relevant items retrievedRecall quantifies how completely a system retrieves relevant data.
Query expansion is used to:Shorten searchesBroaden or refine resultsRemove duplicatesRank by lengthB) Broaden or refine resultsIt adds related terms to improve retrieval accuracy.
Web crawlers primarily:Index web pagesStore passwords Host serversBlock spamA) Index web pagesCrawlers automatically browse and index pages for search engines.
A crawler in a search engine is used to:Rank pagesIndex web pages Analyze grammarCompress dataB) Index web pagesCrawlers scan the web to collect and update indexed pages.
Cosine similarity measures:Difference between vectorsAngle between document vectorsEuclidean distanceBinary overlapB) Angle between document vectorsIt measures similarity based on the cosine of the angle between vectors.
Precision in information retrieval refers to:Relevant results among retrievedTotal retrieved documentsAll relevant documentsData speedA) Relevant results among retrievedPrecision quantifies accuracy of search results.