Building Insights from Unsupervised Data with Machine Learning Meeting 7

lOMoARcPSD|26370826 Unsupervised Application of Machine Learning to Big Data: Quiz 7 Started on State Completed on Time taken Question 1 Correct Marked out of 5.00 Monday, April 19, 2021, 12:02 PM Finished Monday, April 19, 2021, 12:27 PM 25 mins 28 secs A rule-based method for finding relationships between variables in a given data set. This method is often used for market basket analysis, allowing companies to better understand the relationships between different products. Understanding customer consumption habits allows businesses to develop better cross-selling strategies and recommendation engines. Select one: a. Dimensionality reduction b. clustering c. Regression d. Association Question 2 Correct Marked out of 5.00 A type of dimensionality reduction algorithm used to reduce redundancy and compress data sets through feature extraction. This method uses a linear transformation to create a new data representation, resulting in a set of "principal components". The first principal component is the direction that maximizes the variance of the dataset. While the second principal component also finds the maximum variance in the data, it is completely uncorrelated with the first principal component, resulting in a direction that is perpendicular, or orthogonal, to the first component. This process repeats based on the number of dimensions, where the next principal component is direction orthogonal to the previous component with the most variance. Select one: a. Singular value decomposition b. FP-Growth c. Principal component analysis d. Autoencoders Question 3 Correct Marked out of 5.00 A technique that utilizes neural networks to compress data and then recreate a new representation of the original input data. There is a hidden layer that specifically acts as a bottleneck to compress the input layer before reconstructing it in the output layer. Select one: a. Autoencoders b. Principal component analysis c. FP-Growth d. Singular value decomposition 1 Online lOMoARcPSD|26370826 Clustering that takes a "top-down" approach. In this case, a cluster of data is divided based on the differences between data points. Clustering in this way is not commonly used, but is still worth noting in the context of hierarchical clustering. This clustering process is usually visualized using a dendrogram, a tree-like diagram that documents the merging or splitting of data points at each iteration. Select one: a. Overlapping Approach Question 4 b. Agglomerative Approach Correct c. Probabilistic Approach Marked out of 5.00 d. Divisive Approach Question 5 Which is not a task from unsupervised learning Correct Marked out of 5.00 Select one: a. Regression b. clustering c. Dimensionality reduction d. Association Question 6 The most widely used algorithm for Association rules tasks Incorrect Marked out of 5.00 Select one: a. Eclat b. Apriori c. Apriori d. FP-Growth Question 7 Correct Marked out of 5.00 An unsupervised technique that helps solve density estimation or "soft" clustering problems. In this case, data points are clustered based on their likelihood of belonging to a particular distribution. The most commonly used technique is the Gaussian Mixture Model (GMM) Select one: a. Divisive Approach b. Agglomerative Approach c. Probabilistic Approach d. Overlapping Approach Question 8 Correct Marked out of 5.00 A technique used when the number of features, or dimensions, in a given data set is too high. This technique reduces the amount of input data to a manageable size while also maintaining the integrity of the data set as much as possible, usually used in the data preprocessing stage. Select one: a. clustering b. Association c. Dimensionality reduction d. Regression lOMoARcPSD|26370826 An algorithm that attempts to find hidden patterns or groupings in data without the need for human intervention. Select one: a. Semi Supervised Learning Question 9 b. Unsupervised Learning Correct c. Deep Learning Marked out of 5.00 d. Supervised Learning Question 10 Correct Marked out of 5.00 An algorithm that uses a hash tree to enumerate a collection of items, navigating the data set in a first-of-its-kind manner. This algorithm has been popularized through market basket analysis, leading to different recommendation engines for music platforms and online retailers. For example, if I play Black Sabbath radio on Spotify, starting with their song "Orchid," one of the other songs on this channel is likely to be a Led Zeppelin song, such as "Over the Hills and Far Away." This is based on my previous listening habits as well as the habits of others. Select one: a. Apriori b. Apriori c. Eclat d. FP-Growth Question 11 Algorithms that are not used for clustering Correct Marked out of 5.00 Select one: a. Hierarchical b. Auto Encoder c. Overlapping d. Clustering exclusive (specifically exclusive) Question 12 Correct A form of clustering that stipulates that a data point can only exist in one cluster. This can also be referred to as "hard" clustering. One of the algorithms used is K-means clustering. Marked out of 5.00 Select one: a. Hierarchical b. Overlapping c. Exclusive Clustering (specifically exclusive) d. Divisive Question 13 Incorrect Marked out of 5.00 Another dimension reduction approach that factors a matrix, A, into three lower rank matrices. SVD is denoted by the formula, A = USVT, where U and V are orthogonal matrices. S is the diagonal matrix, and the value of S is considered as the single value of matrix A. Principal component analysis Select one: a. FP-Growth b. Singular value decomposition c. Autoencoders lOMoARcPSD|26370826 A technique that groups unlabeled data based on their similarities or differences. These algorithms are used to process raw, unclassified data objects into groups represented by structures or patterns in the information. Select one: a. Dimensionality reduction b. Association Question 14 Correct c. Clustering d. Regression Marked out of 5.00 Question 15 Unsupervised machine learning, using machine learning algorithms to analyze and classify unlabeled data sets Correct Marked out of 5.00 Select one: a. Unsupervised Learning b. Supervised Learning c. Deep Learning d. Semi Supervised Learning Question 16 Correct Clustering as a "bottom-up approach". Where the data points are isolated as separate groupings at first, and then merged together iteratively on the basis of similarity until a single cluster is reached. Marked out of 5.00 Select one: a. Agglomerative Approach b. Overlapping Approach c. Probabilistic Approach d. Divisive Approach Question 17 Correct Its ability to find similarities and differences in information makes unsupervised learning the ideal solution for the following tasks, except Marked out of 5.00 Select one: a. Exploratory data analysis b. Classification of citrus fruits and apples c. Customer segmentation d. cross-selling strategy Question 18 Correct A form of clustering that allows data points to belong to multiple clusters with separate membership degrees. This is also called "soft" or fuzzy k-means clustering. Marked out of 5.00 Select one: a. Exclusive Clustering (specifically exclusive) b. Hierarchical c. Overlapping d. Divisive lOMoARcPSD|26370826 Question 19 The Unsupervised Learning approach is Correct Marked out of 5.00 Select one: a. Learn how unsupervised and supervised learning works b. Learn how deep learning works c. Learning how supervised learning works d. Learning how unsupervised learning works Question 20 Correct Marked out of 5.00 An algorithm that uses a hash tree to enumerate a collection of items, navigating the data set in a first-of-its-kind manner. This algorithm has been popularized through market basket analysis, leading to different recommendation engines for music platforms and online retailers. For example, if I play Black Sabbath radio on Spotify, starting with their song "Orchid," one of the other songs on this channel is likely to be a Led Zeppelin song, such as "Over the Hills and Far Away." This is based on my previous listening habits as well as those of others. This includes tasks Select one: a. Regression b. Association c. Dimensionality reduction d. clustering Jump to...