1. (Choose 1 answer)
What is the primary content of most documents represented?
A. Metadata
B. Images
C. Words
D. Graphics
Đáp án:

2. (Choose 1 answer)
What term is commonly used for a set of web pages that are all about the same topic?
A. Internet ensemble
B. Digital assortment
C. Web consortium
D. Web community
Đáp án:

3. (Choose 1 answer)
Which of the following correctly identifies the on-page and off-page factor groups?
A. Content, Links, Trust are on-page factors; Architecture, HTML, User are off-page factors
B. Content, Architecture, HTML are on-page factors; Trust, Links, User are off-page factors
C. Content, Likes, User are on-page factors; Trust, Links, Architecture are off-page factors
D. Content, Links, User are on-page factors; Trust, HTML, Architecture are off-page factors
Đáp án:

4. (Choose 1 answer)
Mean Average Precision (MAP) is defined as:
A. Average of DCG across queries
B. Mean of per-query average precision values
C. Precision at rank k averaged
D. Area under the ROC curve
Đáp án:

5. (Choose 1 answer)
What is the technique used for recognizing each word occurrence in a document?
A. Tokenizing
B. Lexical analysis
C. Metadata
D. Markup languages
Đáp án:

6. (Choose 2 answers)
Why do web crawlers use politeness policies? Check all correct answers.
A. To be respectful of web servers and avoid overloading them
B. To improve the efficiency of the web
C. To avoid getting banned from websites
D. To speed up crawling web
Đáp án:

7. (Choose 1 answer)
Cosine similarity is usually computed between:
A. Raw term counts
B. Binary vectors
C. L2-normalized TF-IDF vectors
D. PageRank values
Đáp án:

8. (Choose 1 answer)
What is the primary difference between on-page factors and off-page factors in search engine optimization?
A. On-page factors involve external elements like links and likes
B. On-page factors are related to website content, while off-page factors are not
C. Off-page factors can be directly controlled on one's website, while on-page factors cannot
D. On-page factors primarily include Trust, Links, and User groups
Đáp án:

9. (Choose 1 answer)
Which of the following is NOT a benefit of relevance feedback?
A. It allows users to refine their search results without having to enter a new query
B. It can improve the accuracy of search results by helping the system to better understand the user's intent
C. It is a fully automated process that does not require any user input
D. It can be used to identify relevant documents in a ranking based on the initial query
Đáp án:

10. (Choose 1 answer)
What is NOT the reason for a query being a poor representation of the information need?
A. The user is encouraged to enter short queries
B. The user finds it difficult to express the information need
C. The search engine interface is not user-friendly
D. Long queries often fail
Đáp án:

11. (Choose 1 answer)
Why is on-page SEO considered crucial for better rankings in search engine results?
A. It primarily involves optimizing external factors that influence rankings
B. It helps search engines understand the content and purpose of a page
C. On-page SEO is not important for search engine rankings
D. It focuses on offline marketing efforts, which directly impact rankings
Đáp án:

12. (Choose 1 answer)
What is the task of search engine related to determine the document relevance?
A. Ranking
B. Testing
C. Measuring
D. Creating user interaction
Đáp án:

13. (Choose 1 answer)
When is recall not an appropriate measure in search applications, according to the text?
A. When users extensively explore lower-ranked results
B. When tasks involve multiple relevant documents
C. When users focus on the first page or two of results
D. When navigational search is the primary task
Đáp án:

14. (Choose 1 answer)
In IR, the set of documents we search over is commonly called the:
A. Ledger
B. Registry
C. Corpus
D. Cache
Đáp án:

15. (Choose 1 answer)
What is a fundamental concept in information retrieval?
A. Relevance
B. Completeness
C. Scalability
D. Stability
Đáp án:

16. (Choose 1 answer)
What is the primary consideration when optimizing content for search engine optimization?
A. Ensuring a high word count in the documents
B. Including as many keywords as possible in the text
C. Ensuring the quality of content and error-free writing
D. Optimizing for rarely used queries exclusively
Đáp án:

17. (Choose 2 answers)
Which fields are standard in anonymized query logs? (Choose 2)
A. Query timestamp
B. Result URLs shown
C. User passwords
D. Credit card numbers
Đáp án:

18. (Choose 1 answer)
What distinguishes social media tagging, as seen in folksonomies, from the approach of the semantic web?
A. Social media tagging uses rigid, predefined ontologies, while the semantic web is based on user-driven folksonomies
B. Social media tagging relies on standardized, fixed ontologies, while the semantic web allows a relatively unrestricted set of tags meaningful to users
C. Social media tagging is primarily used for organizing information, while the semantic web focuses on sharing information
D. Social media tagging is restricted to a small number of web pages, while the semantic web is widely adopted
Đáp án:

19. (Choose 1 answer)
What is the purpose of the text transformation component in HTML and XML documents?
A. To remove control sequences and non-content data
B. To convert the documents to a consistent text format
C. To extract metadata from the documents
D. To prepare the documents for further processing
Đáp án:

20. (Choose 1 answer)
Why is evaluating the output of a clustering algorithm challenging, especially in the absence of labeled data?
A. Because clustering algorithms do not produce any output
B. Because there is no way to assess the quality of the clusters
C. Because the evaluation metric depends on the algorithm, and vice versa
D. Because clusters are assigned explicit and meaningful labels
Đáp án:

21. (Choose 1 answer)
What is the score for a document in document-at-a-time retrieval?
A. The sum of the word counts in each document
B. The number of words in each document
C. The number of documents containing each word
D. The number of times each word appears in each document
Đáp án:

22. (Choose 1 answer)
What are the major components of the indexing process in search engine components?
A. Text acquisition, text transformation, and index creation
B. Text acquisition, text transformation, and query processing
C. Text acquisition, query processing, and index creation
D. Text transformation, query processing, and index creation
Đáp án:

23. (Choose 1 answer)
Which of the following is NOT a major goal of contextual advertising?
A. To find advertisements that are relevant to the query context
B. To increase the number of user clicks on advertisements
C. To generate revenue for search engine companies
D. To provide users with information about products and services that they may be interested in
Đáp án:

24. (Choose 1 answer)
How are user tags typically used on social media websites?
A. User tags are automatically assigned to items by the website
B. User tags are single-term labels applied to user-created content
C. User tags are primarily used for visualizing clusters of content
D. User tags are used for manual indexing, allowing users to represent content with assigned terms
Đáp án:

25. (Choose 1 answer)
What is the main assumption of Naive Bayes classification?
A. Equal class priors
B. Conditional independence of features
C. Linear separation
D. Zero variance features
Đáp án:

26. (Choose 1 answer)
Which language typically requires segmentation in tokenization because words are not whitespace-delimited?
A. English
B. Vietnamese
C. Chinese
D. German
Đáp án:

27. (Choose 1 answer)
What is a difference between agglomerative clustering and the K-means algorithm regarding the cluster assignments of instances?
A. Instances within the same cluster may end up in different clusters as the algorithm proceeds in K-means algorithm but in agglomerative clustering
B. Instances are always assigned to the same cluster in agglomerative clustering but in K-means algorithm
C. Instances can never be assigned to different clusters in K-means algorithm but in in agglomerative clustering
D. Instances can change clusters during each iteration in agglomerative clustering but in K-means algorithm
Đáp án:

28. (Choose 1 answer)
What is the primary advantage of conjunctive processing to short queries?
A. It improves effectiveness and efficiency simultaneously
B. It is more flexible than other query optimization techniques
C. It is more accurate than other query optimization techniques
D. It is more efficient than other query optimization techniques
Đáp án:

29. (Choose 1 answer)
Query expansion techniques are usually based on an analysis of word or term co-occurrence in
A. The small fraction of the entire document collection
B. A large collection of queries
C. The top-ranked documents in a result list
D. A large collection of queries and the top-ranked documents in a result list
Đáp án:

30. (Choose 1 answer)
What is the biggest challenge of crawling web pages?
A. The sheer scale of the Web
B. The cost of crawling the Web
C. The technical challenges of crawling the Web
D. The security of the Web
Đáp án:

31. (Choose 1 answer)
What is the primary difference between effectiveness and efficiency in search engines evaluation?
A. Effectiveness measures how quickly the search engine retrieves information, while efficiency gauges the relevance of the results
B. Effectiveness evaluates the search engine's ability to find relevant information, while efficiency assesses how quickly it accomplishes this
C. Effectiveness measures the time and space requirements of the algorithm, while efficiency evaluates how well the ranking corresponds to user judgments
D. Effectiveness and efficiency are synonymous terms used interchangeably in the evaluation of search engines
Đáp án:

32. (Choose 1 answer)
What is the minimum score that each document needs to reach before it can be shown to the user?
A. The score of the highest scoring document
B. The score of the lowest scoring document
C. The score of the kth-highest scoring document
D. The score of the kth-lowest scoring document
Đáp án:

33. (Choose 2 answers)
Which features are typically considered query-independent (static)? (Choose 2)
A. Page age
B. BM25 score
C. URL depth
D. Query length
Đáp án:

34. (Choose 1 answer)
What are the two major functions supported by search engine components?
A. Indexing process and query process
B. Crawling process and ranking process
C. Indexing process and ranking process
D. Crawling process and query process
Đáp án:

35. (Choose 1 answer)
What is the focus of search engine optimizers in the context of search engine ranking factors?
A. Creating websites that are successful in the short term changes
B. Reacting to short-term changes in single ranking factors
C. Ignoring search engine updates for a pragmatic approach
D. Focusing exclusively on basic procedures for achieving higher ranking
Đáp án:

36. (Choose 1 answer)
What is the purpose of the parsing component in information retrieval systems?
A. To identify structural elements in documents, such as titles, figures, links, and headings
B. To convert documents and queries into a consistent text format
C. To rank documents based on their relevance to a query
D. To store documents in a way that is efficient for searching
Đáp án:

37. (Choose 3 answers)
What is a character encoding? (3 options)
A. A way to represent text using a series of bits
B. A way to map bits to glyphs
C. A way to make HTML files compatible with each other
D. A way to translate web page text into images for better display
Đáp án:

38. (Choose 1 answer)
What is a notable drawback associated with using query logs for web search engine development?
A. They are not accessible to researchers
B. They provide more precise relevance judgments than traditional methods
C. They represent a smaller user and query sample
D. They lack the same level of precision as explicit relevance judgments
Đáp án:

39. (Choose 1 answer)
Which of the following is NOT a query expansion technique?
A. Finding synonyms of words
B. Adding typical words to query
C. Stemming words
D. Using a general thesaurus
Đáp án:

40. (Choose 1 answer)
What is the key distinction between card catalog generation and social media tagging?
A. Card catalogs are generated by non-experts, while social media tagging is done by experts
B. Card catalogs use user-generated ontologies, while social media tagging relies on controlled vocabularies
C. Card catalogs ensure standardized descriptors, while user tags in social media are not standardized
D. Card catalogs form their own descriptions of concepts, while social media tagging uses fixed vocabularies
Đáp án:

41. (Choose 1 answer)
How do backlinks contribute to building authority for a website?
A. They directly increase your website's conversion rate
B. They make your content more complex and challenging
C. They signal to search engines that your website is an authority
D. They reduce the website's visibility on search engine result pages
Đáp án:

42. (Choose 1 answer)
Is there a trade-off between efficiency and effectiveness in search engine design?
A. Yes, there is recently a trade-off, and search engine designers often have to choose between one or the other
B. No, search engines can simultaneously maximize both efficiency and effectiveness without any compromise
C. Yes, but current search engine techniques prioritize effectiveness over efficiency
D. No, efficiency considerations are not relevant to search engine design
Đáp án:

43. (Choose 1 answer)
What structure accelerates merging postings by allowing jumps within a long list?
A. Bloom filters
B. Skip pointers
C. Suffix arrays
D. B-trees
Đáp án:

44. (Choose 1 answer)
What is the main difference between document distribution and term distribution?
A. In document distribution, each index server stores a different part of the inverted index, while in term distribution, each index server stores a copy of the entire document collection
B. In document distribution, the director is responsible for merging the results from the index servers, while in term distribution, one of the index servers is responsible for processing the query and merging the results from the other index servers
C. Document distribution is more difficult to implement than term distribution
D. seek time for document distribution is shorter than term distribution with the same number of terms queried over the same number of index servers
Đáp án:

45. (Choose 1 answer)
What is the purpose of a port in a network?
A. To translate a hostname into an IP address
B. To identify a particular service on a server computer
C. To store web pages that have been downloaded
D. To send requests for web pages to a server computer
Đáp án:

46. (Choose 1 answer)
Which of the following is NOT true?
A. When calculating the similarity between Document and Query, Euclidean similarity is worse than cosine similarity because normally, the query vector is more likely to be shorter than the document vector
B. When calculating the similarity between Document and Query, Jaccard similarity is better than Inner similarity because Inner similarity penalizes the dissimilarity
C. Jaccard coefficient separates the scores of relevant documents further apart
D. The inner similarity between two large documents tends to be larger the inner similarity between two small documents
Đáp án:

47. (Choose 1 answer)
Which feedback method moves the query vector closer to relevant documents?
A. Rocchio
B. HITS
C. PageRank
D. AdaRank
Đáp án:

48. (Choose 1 answer)
What is the primary goal of the semantic web?
A. The semantic web aims to restrict the use of metadata tags on web content
B. The semantic web intends to limit user-driven tagging on the web
C. The semantic web seeks to develop and consistently use standardized metadata tags
D. The semantic web aims to prioritize user-driven folksonomies over rigid ontologies
Đáp án:

49. (Choose 1 answer)
How are recommender systems different from document filtering systems?
A. Recommender systems retrieve relevant documents based on long-term information needs, while document filtering systems predict user ratings for items
B. Recommender systems are used for standard web search, while document filtering systems are not
C. Recommender systems focus on short-term information needs, while document filtering systems focus on long-term information needs
D. Recommender systems predict user item ratings and make recommendations, while document filtering systems monitor incoming documents
Đáp án:

50. (Choose 1 answer)
What is the difference between a static web page and a dynamic web page?
A. A static web page is stored on a web server and displayed in a web browser unmodified, while a dynamic web page is generated by code executing on the web server or the client
B. A static web page is only accessible through web forms, while a dynamic web page can be accessed directly
C. A static web page is more difficult to crawl than a dynamic web page
D. A static web page is impossible to crawl and a dynamic web page is only accessed through web forms
Đáp án:

Đang tải dữ liệu

Đang tải dữ liệu

Đang tải dữ liệu