Review Sheet 1 Question I: MCQ 1) Which of the following refers to the problem of finding abstracted patterns (or structures) in the unlabeled data? a. b. c. d. Supervised learning Unsupervised learning Hybrid learning Reinforcement learning 2) Which one of the following refers to querying the unstructured textual data? a. b. c. d. Information access Information update Information retrieval Information manipulation 3) Which of the following can be considered as the correct process of Data Mining? a. b. c. d. Infrastructure, Exploration, Analysis, Interpretation, Exploitation Exploration, Infrastructure, Analysis, Interpretation, Exploitation Exploration, Infrastructure, Interpretation, Analysis, Exploitation Exploration, Infrastructure, Analysis, Exploitation, Interpretation 4) Which of the following is an essential process in which the intelligent methods are applied to extract data patterns? a. b. c. d. Warehousing Data Mining Text Mining Data Selection 5) What is KDD in data mining? a. b. c. d. Knowledge Discovery Database Knowledge Discovery Data Knowledge Data definition Knowledge data house 6) The adaptive system management refers to: a. b. c. d. Science of making machine performs the task that would require intelligence when performed by humans. A computational procedure that takes some values as input and produces some values as the output. It uses machine learning techniques, in which programs learn from their past experience and adapt themself to new conditions or situations. All of the above. 7) For what purpose, the analysis tools pre-compute the summaries of the huge amount of data? a. b. c. d. In order to maintain consistency For authentication For data access To obtain the queries response 8) What are the functions of Data Mining? a. b. c. d. Association and correctional analysis classification Prediction and characterization Cluster analysis and Evolution analysis All of the above 9) In the following given diagram, which type of clustering is used? a. b. c. d. Hierarchal Naive Bayes Partitional None of the above 10) Which of the following statements is incorrect about the hierarchal clustering? a. b. c. d. The hierarchal type of clustering is also known as the HCA The choice of an appropriate metric can influence the shape of the cluster In general, the splits and merges both are determined in a greedy manner All of the above 11) Which one of the following can be considered as the final output of the hierarchal type of clustering? a. b. c. d. A tree which displays how the close thing are to each other Assignment of each point to clusters Finalize estimation of cluster centroids None of the above 12) Which one of the following statements about the K-means clustering is incorrect? a. b. c. d. The goal of the k-means clustering is to partition (n) observation into (k) clusters K-means clustering can be defined as the method of quantization The nearest neighbor is the same as the K-means All of the above 13) Which of the following technique is supervised learning? a. b. Classification Clustering c. d. Association rules None of the above 14) Which one of the clustering technique needs the merging approach? a. b. c. d. Partitioned Naïve Bayes Hierarchical Both A and C 15) The self-organizing maps can also be considered as the instance of _________ type of learning. a. b. c. d. Supervised learning Unsupervised learning Missing data imputation Both A & C 16) The following given statement can be considered as the examples of_________. “Suppose one wants to predict the number of newborns according to the size of storks' population by performing supervised learning” a. b. c. d. Structural equation modeling Clustering Regression Classification 17) In the example predicting the number of newborns, the final number of total newborns can be considered as the _________ a. b. c. d. Features Observation Attribute Outcome 18) Which of the following statement best describes classification? a. b. c. d. It is a measure of accuracy It is a function that maps a data object to a predefined class It is the task of assigning a classification None of the above 19) Which of the following statements is correct about data mining? a. b. c. d. It can be referred to as the procedure of mining knowledge from data Data mining can be defined as the procedure of extracting information from a set of the data The procedure of data mining also involves several other processes like data cleaning, data transformation, and data integration All of the above 20) In data mining, how many categories of functions are included? a. b. 5 4 c. d. 2 3 21) Which of the following can be considered as mapping of a set or objects with some predefined group or classes? a. b. c. d. Data set Data Characterization Data Sub Structure Data Discrimination 22) The analysis performed to uncover the interesting statistical relationship between associated -attributes value pairs are known as the _______. a. b. c. d. Mining of association Mining of correlation Mining of clusters All of the above 23) Which one of the following can be defined as the data object which does not comply with the general behavior (or the model of available data)? a. b. c. d. Evaluation Analysis Outliner Analysis Classification Prediction 24) Which one of the following statements is correct about the data cleaning? a. b. c. d. It refers to removing noise and duplicate data It refers to the transformation of wrong data into correct data It refers to removing missing and inconsistent data All of the above 25) The classification of the data mining system involves: a. b. c. d. Database technology Information Science Machine learning All of the above 26) The issues like efficiency, scalability of data mining algorithms comes under_______ a. b. c. d. Performance issues Diverse data type issues Mining methodology and user interaction All of the above 27) Which one of the following correctly defines the term cluster? a. b. Group of similar objects that differ significantly from other objects Symbolic representation of facts or ideas from which information can potentially be extracted c. d. Operations on a database to transform or simplify data in order to prepare it for a machine-learning algorithm All of the above 28) Which one of the following refers to the binary attribute? a. b. c. d. This takes only two values. In general, these values will be 0 and 1, and they can be coded as one bit The natural environment of a certain species Systems that can be used without knowledge of internal operations All of the above 29) Which of the following correctly refers the data selection? a. b. c. d. A subject-oriented integrated time-variant non-volatile collection of data in support of management The actual discovery phase of a knowledge discovery process The stage of selecting the right data for a KDD process All of the above 30) Which one of the following correctly refers to the task of the classification? a. b. c. d. A measure of the accuracy, of the classification of a concept that is given by a certain theory The task of assigning a classification to a set of examples A subdivision of a set of examples into a number of classes None of the above 31) Which of the following correctly defines the term "Hybrid"? a. b. c. d. Approach to the design of learning algorithms that is structured along the lines of the theory of evolution. Decision support systems that contain an information base filled with the knowledge of an expert formulated in terms of if-then rules. Combining different types of method or information None of these 32) Which of the following correctly defines the term "Discovery"? a. b. c. d. It is hidden within a database and can only be recovered if one is given certain clues (an example IS encrypted information). An extremely complex molecule that occurs in human chromosomes and that carries genetic information in the form of genes. It is a kind of process of executing implicit, previously unknown and potentially useful information from data None of the above 33) Euclidean distance measure is defined as ___________ a. b. c. d. The process of finding a solution for a problem simply by enumerating all possible solutions according to some predefined order and then testing them The distance between two points as calculated using the Pythagoras theorem A stage of the KDD process in which new data is added to the existing selection. All of the above 34) Which one of the following can be considered as the correct application of the data mining? a. b. c. d. Fraud detection Corporate Analysis & Risk management Management and market analysis All of the above 35) Which of the following refers to the sequence of pattern that occurs frequently? a. b. c. d. Frequent sub-sequence Frequent sub-structure Frequent sub-items All of the above 36) The issues like "handling the rational and complex types of data" comes under which of the following category? a. b. c. d. Diverse Data Type Mining methodology and user interaction Issues Performance issues All of the above 37) Which of the following also used as the first step in the knowledge discovery process? a. b. c. d. Data selection Data cleaning Data transformation Data integration 38) Which of the following refers to the steps of the knowledge discovery process, in which the several data sources are combined? a. b. c. d. Data selection Data cleaning Data transformation Data integration 39) In certain cases, it is not clear what kind of pattern need to find, data mining should_________: a. b. c. d. Try to perform all possible tasks Perform both predictive and descriptive task It may allow interaction with the user so that he can guide the mining process All of the above 40) Which of the following is a probabilistic model? a. b. c. d. Decision tree Naïve bayes Artificial Neural Network Support Vector Machine 41) An example of continuous attribute is a. b. c. d. Color Gender Weight Zip code 42) An example of discrete attribute is a. b. c. d. Temperature Height Income Marital status 43) An attribute encompassing the notion of Good, Better, Best is a. b. c. d. Nominal Ordinal Interval Ratio 44) The Celsius temperature is --------- attribute a. b. c. d. Nominal Ordinal Interval Ratio 45) The length is --------- attribute a. b. c. d. Nominal Ordinal Interval Ratio 46) A categorical attribute can be a. b. c. d. Nominal or ordinal Ordinal or Interval Interval or ratio Ratio or nominal 47) which of the following operations are applicable to Nominal attribute a. b. c. d. Distinctness Addition Multiplication Order 48) Which of the following operations are applicable to Ratio attribute a. Distinctness b. Addition/Subtraction c. Multiplication/Division d. All of the above 49) Which of the following operations are applicable to temperature in Fahrenheit a. b. c. d. Division Addition Multiplication All of the above 50) Which of the following are ordered data a. b. c. d. Temporal data Molecular Structures Document Data World Wide Web 51) The data in the table are what type of data a. b. c. d. Spatial Data Transaction Data Document Data Sequential Data TID Items 1 2 3 4 5 Bread, Coke, Milk Beer, Bread Beer, Coke, Diaper, Milk Beer, Bread, Diaper, Milk Coke, Diaper, Milk Question II: Fill in the spaces 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. social media, cyber security, satellite sensors Sources of Big Data includes -------,-------, and ------5 Vs Characteristics of big data includes --------, -------, --------, --------, and --------Velocity High speed data generation and storage is called ------semi-structured, unstructured Variety means different types of data such as --------, --------, and Structured, -------Examples of high speed data are -------, -------, -------, and ---------------- is the Non-trivial extraction of implicit, previously unknown and potentially useful knowledge from data Datamining Ai, Machine learning, statistics, Datamining draws ideas from other fields such as ------, -------, ------, ------, and ------database, pattern recognition The main phases in datamining are -------,-------,-------, -------, and -------Aggregation, sampling, feature selection, dimensionality Data preprocessing comprises ------,-------,------,and ------reduction Data cleaning involves --------, -------, and ------Datamining techniques includes two categories, they are --------- andpredictive, ----------- descriptive supervised In datamining there are two types of learning, they are --------- and (un) -------Classification is and example for -------- learning while ------- is unsupervised learning supervised, clustering Association rule is a type of -------- learning, while regression is -------- learning Supervised learning creates models from ---------- data objects labeled 16. Unsupervised learning creates models from ---------- data objects unlabeled classes 17. A classifier maps input data objects topredefined ---------18. --------- finds groups of objects such that the objects in a group will be similar to one another and different from the objects in other groups clustering 19. -------- use some variables to predict unknown or future values of other variables predictive method 20. -------- find human-interpretable patterns that describe the data descriptive 21. Sky Survey Cataloging is an application of classification -------clustering 22. Market segmentation is an application of ------23. --------- predicts a value of a given continuous valued variable based on the values of other variables, assuming a linear or nonlinear model of dependency regression 24. Given a set of records each of which contain some number of items from a given collection ---------- produce dependency rules which will predict occurrence of an item based on occurrences of other items. Association rules 25. ------- is an application of association rules market-basket analysis 26. ------- detects significant deviations from normal behavior anomaly detection anomaly detection 27. Network Intrusion Detection is an application of --------28. Examples of classification models are ------, -------, and KNN, ------- naïve Bayes, neural network, tree 29. Examples of clustering models are --------, ---------, and -------- hieratical, density, grid, model based 30. Web mining employs -------- technique. clustering 31. An ----------- is a property or characteristic of an object attribute object 32. A collection of attributes describes an ………. field, feature, dimension, characteristic 33. Attribute is also known as …….., ……., or ----------34. Object is also known as …………, ………..,, or ---------- record, instance, sample 35. Types of attributes are ----------,------------,------------, and --------- nominal, ordinal, interval, ratio 36. ----------- attribute has only a finite or countably infinite set of values discrete 37. ----------- attribute has real numbers as attribute values continuous sparsity, resolution, size 38. The important characteristics of data include -------,----------, ----------, anddimensionality, -----------noise, outliers, fake, missing, wrong, duplicate 39. Example of problem that affect data quality are ……….,…………….,………….,and……… 40. ------------ are data objects with characteristics that are considerably different than most of the other data objects in the data set outliers Question III: Problems 1. Assume X and Y are two object in a training dataset where X= [ 2 4 5 -1 0] and Y=[1 4 -3 6 7] Compute the Euclidean Distance between X and Y 2. If X and Y are two binary objects where X= 1000000000 Y= 0000001001 Compute the proximity using Jacard and Simple Matching Coefficients 3. Compute the similarity between the two documents d1 = 3 2 0 5 0 0 0 2 0 0 d2 = 1 0 0 0 0 0 0 1 0 2 4. Compute the correlation between the following two objects x = (1, 2, 4, 3, 0, 0, 0), y = (1, 2, 3, 4, 0, 0, 0) 5. For the following splits compute a. Gini-split b. Gain-split c. Gain-ratio d. Classification error Parent C0 7 C1 5 Node 1 Node 2 C0 5 2 C1 1 4 Question IV: True/False 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Ordinal attributes are considered as continuous attributes false Classification is a type of unsupervised learning false Nominal attributes are quantitative false Data cleaning comprises removing noise and duplicated data true A classification model is induced from unlabeled data false true A cluster is a group of similar objects Recommendation System predicts whether a user would like a given resource or not true Document recognition can recognize whether an e-mail is spam or not, based on the message header and content true Choosing a model is the first step in pattern recognition life cycle false Clustering Model is trained with data samples with unknown labels true Number of classes is known a priori true Entropy measure is used for evaluating proximity false Gini index can be used as a measure for impurity measure true true SVM is an algorithm for clustering K-means is an algorithm for classification false ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
