Uploaded bySimplilearn
Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners Part - 2 | Simplilearn
The document discusses clustering, specifically k-means clustering, explaining its purpose as an unsupervised learning method for grouping unlabeled data based on feature similarity. It also covers logistic regression, a classification algorithm for binary and multi-classification problems, highlighting its use in predicting outcomes like the malignancy of tumors. Examples and methods such as the elbow method for determining the number of clusters in k-means are also provided.
In this document
Powered by AI
Introduces clustering, defines it as organizing objects into groups based on similarity, and offers examples like K-Means clustering and organizing books.
Explains K-Means clustering as an unsupervised learning method used for unlabeled data to form clusters based on feature similarity.
Details the steps for K-Means clustering including initialization of centroids, distance computation, and cluster assignments until stabilization.
Introduces a flowchart to visualize the K-Means clustering process from choosing K to determining stable clusters.
Presents a dataset for K-Means clustering and the method for selecting initial centroids for the clustering process.
Describes how data points are assigned to clusters, centroid recalculations, and stabilizing clusters with final outputs.
Discusses using the elbow method to select optimal clusters and provides a use case of K-Means clustering for car brands.
Begins the discussion on logistic regression, clarifying its purpose for classification problems and contrasting it with linear regression.
Explains why logistic regression is crucial for categorical outcomes and introduces the sigmoid function in the context of binary outcomes.
Illustrates a practical logistic regression use case for classifying tumors, achieving a prediction accuracy of 91%.
Explores the concept of data measures and dimensions, providing examples of clustering, classification, and regression use cases.
Summarizes the K-Means elbow method for clustering cars and logistic regression for tumor classification.
More Related Content
Similar to Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners Part - 2 | Simplilearn
More from Simplilearn
Machine Learning Tutorial Part - 2 | Machine Learning Tutorial For Beginners Part - 2 | Simplilearn
- 1. Clustering What is Clustering? K-MeansClustering Flowchart to understand K-means Clustering Clustering of cars based on brands Logistic Regression What’s in it for you? What is Logistic Regression? Logistic Regression Curve & Sigmoid function Classify whether a tumor is malignant or benign based on features Cover/transition slides will be changed
- 2. Clustering Suppose, we have apile of books of different genres!
- 3. Clustering Now, we dividethem into different groups like Fiction Horror Educational
- 4. Well, organizing objects intogroups based on their similarity is Clustering!
- 5. Well, organizing objects intogroups based on their similarity is Clustering! K-means Clustering
- 6. K-Means Clustering isan example of Unsupervised learning
- 7. K-Means Clustering isan example of Unsupervised learning It is used when you have unlabeled data!
- 8. K-Means Clustering isan example of Unsupervised learning It is used when you have unlabeled data! To find clusters in the data based on feature similarity!
- 9. Steps for K-Means Supposewe have these data points and we want to assign them into clusters
- 10. STEP 1: InitializeCluster Centroids We pick ‘K’ clusters & assign random centroids to clusters
- 11. STEP 1: InitializeCluster Centroids We pick ‘K’ clusters & assign random centroids to clusters Then, we compute distance from objects to centroids
- 12. STEP 2: ComputeMinimum Distance Now, we form new clusters based on minimum distance and calculate their centroids
- 13. STEP 3: AssignPoints to New Clusters Repeat previous two steps iteratively till the cluster centroids stop changing their positions and become static
- 14. STEP 3: AssignPoints to New Clusters Repeat previous two steps iteratively till the cluster centroids stop changing their positions and become static
- 15. Shall we seea flowchart to understand?
- 16. Flowchart to understandK-Means Choose K (Elbow Method) START Assign random centroids to clusters Compute distance from objects to centroids Yes Form new clusters based on minimum distance and calculate their centroids Compute distance from objects to new centroids Repeat until no observations change groups
- 17. Let’s see anexample!
- 18. K-Means Algorithm Subject AB 1 1 1 2 1.5 2 3 3 4 4 5 7 5 3.5 5 6 4.5 5 7 3.5 4.5 Suppose, we have this dataset of 7 individuals and their score on two topics (A and B)
- 19. K-Means Algorithm Now, letstake two farthest-apart points as initial cluster centroids Subject A B 1 1 1 2 1.5 2 3 3 4 4 5 7 5 3.5 5 6 4.5 5 7 3.5 4.5
- 20. K-Means Algorithm Now, letstake two farthest-apart points as initial cluster centroids
- 21. K-Means Algorithm Each pointis then assigned to the closest cluster with respect to their distance from the centroids Cluster 1 Cluster 2
- 22. K-Means Algorithm Now, weagain calculate the centroids of each cluster: Individual Mean Vector (centroid) Cluster 1 1, 2, 3 (1.8, 2.3) Cluster 2 4, 5, 6, 7 (4.1, 5.4) Cluster 1 Cluster 2
- 23. K-Means Algorithm We compareeach individual’s distance to its own cluster mean and to that of the opposite cluster. And we find: Individual Distance to mean (centroid) of Cluster 1 Distance to mean (centroid) of Cluster 2 1 1.5 5.4 2 0.4 4.3 3 2.1 1.8 4 5.7 1.8 5 3.2 0.7 6 3.8 0.6 7 2.8 1.1 Using Eucledian Distance between the points and the mean Cluster 1 Cluster 2
- 24. K-Means Algorithm Individual Distance tomean (centroid) of Cluster 1 Distance to mean (centroid) of Cluster 2 1 1.5 5.4 2 0.4 4.3 3 2.1 1.8 4 5.7 1.8 5 3.2 0.7 6 3.8 0.6 7 2.8 1.1 Only individual 3 is nearer to the mean of the opposite cluster (Cluster 2) than its own (Cluster 1) Cluster 1 Cluster 2 Moving point 3 to new cluster
- 25. K-Means Algorithm Thus, individual3 is relocated to Cluster 2 resulting in the new partition: Individual Distance to mean (centroid) of Cluster 1 Distance to mean (centroid) of Cluster 2 1 1.5 5.4 2 0.4 4.3 3 2.1 1.8 4 5.7 1.8 5 3.2 0.7 6 3.8 0.6 7 2.8 1.1 Cluster 1 Cluster 2
- 26. K-Means Algorithm For thenew clusters, we will find the actual cluster centroids: Individual Mean Vector (centroid) Cluster 1 1, 2, 3 (1.25, 1.5) Cluster 2 4, 5, 6, 7 (3.9, 5.1) Cluster 1 Cluster 2
- 27. K-Means Algorithm On comparingthe distance of each individual’s distance to it’s own cluster mean and to that of the opposite cluster, we find that the data points are stable, hence we have our final clusters! Cluster 1 Cluster 2
- 28. K-Means Algorithm To findappropriate number of clusters in a dataset, we use elbow method: WSS No . of. clusters Elbow point Within sum of squares (WSS) is defined as the sum of the squared distance between each member of the cluster and its centroid Finding the optimal number of clusters using the elbow of the graph is called as the Elbow method
- 29. Use Case Using K-meansclustering to cluster cars into brands using the parameters such as horsepower, cubic inches, make year, etc. Dataset: Cars data having information about 3 brands of cars namely Toyota, Honda, Nissan
- 30.
- 31.
- 32.
- 33.
- 34.
- 35.
- 36.
- 37.
- 38.
- 39. Clustering Today, we’ll diveinto K- means Clustering! Well, organizing objects into groups based on their similarity is Clustering! Logistic Regression
- 40. Logistic Regression Now, let’slook into Logistic Regression
- 41. Logistic Regression The LogisticRegression algorithm is the simplest classification algorithm used for binary or multi-classification problems
- 42. Logistic Regression To brushup, y = mx+c The dependent variable is the target class variable we are going to predict In the previous tutorial, we learnt about Linear Regression, dependent and independent variables
- 43. Logistic Regression In theprevious tutorial, we learnt about Linear Regression, dependent and independent variables The independent variables (x1…xn) are the features or attributes we are going to use to predict the target class To brush up, y = mx+c The dependent variable is the target class variable we are going to predict
- 44. Logistic Regression 1 0 Marks No. ofhours studied We know what a linear regression looks like, but using this graph we cannot divide the outcome into categories 100
- 45. Logistic Regression 100 0 We knowwhat a linear regression looks like, but using this graph we cannot divide the outcome into categories For example, a linear regression graph can tell us that with increase in number of hours studied, the marks of a student will increase But, it will not tell us whether the student will pass or not! Marks No. of hours studied
- 46. Logistic Regression In suchcases, where we need the output as categorical value, we will use logistic regression! 100 0 No. of hours studied Marks
- 47. Logistic Regression 0 100 1 0 Sigmoid Curve SigmoidFunction y = m*x + c p = 1 1 + ⅇ − y p ln ( 1-p ) = m*x + c No. of hours studied No. of hours studied Marks Marks
- 48. Logistic Regression 0 0.2 0.4 0.6 0.8 1 1.2 0 12 3 4 5 6 7 8 9 Logistic Regression Threshold value Probability > 0.50 Value is rounded off to 1 indicating that the student will pass Probability < 0.50 , the value is rounded off to 0 indicating that the student will fail 0.30 0.82
- 49. Problem statement: Toclassify whether a tumor is ‘malignant’ or ‘benign’
- 50.
- 51.
- 52.
- 53.
- 54.
- 55.
- 56.
- 57.
- 58.
- 59. Use Case So, thismodel is able to predict the type of tumor with 91% accuracy!
- 60. Finally, let’s discussthe answers to the quiz asked in Machine Learning Tutorial Part-1 for the instructor
- 61. What do youunderstand from Measures and Dimensions? Each field from the data source is automatically assigned a datatype (such as string, integer) and a role (dimension or measure) Aggregation applied on measures is ‘Sum’ by default but you can always change the default aggregation in the settings Can you tell what’s happening in the following cases? A. Grouping documents into different categories based on the topic and content of each document “This is an example of Clustering where K-means clustering can be used to group the documents by topics using bag-of-words approach”
- 62. What do youunderstand from Measures and Dimensions? Each field from the data source is automatically assigned a datatype (such as string, integer) and a role (dimension or measure) Aggregation applied on measures is ‘Sum’ by default but you can always change the default aggregation in the settings Can you tell what’s happening in the following cases? B. Identifying hand-written digits in images correctly “This is an example of Classification. The traditional approach to solving this would be to extract digit dependent features like curvature of different digits, etc. and then use a classifier like SVM to distinguish between images”
- 63. What do youunderstand from Measures and Dimensions? Each field from the data source is automatically assigned a datatype (such as string, integer) and a role (dimension or measure) Aggregation applied on measures is ‘Sum’ by default but you can always change the default aggregation in the settings Can you tell what’s happening in the following cases? C. Behavior of a website indicating that the site is not working as designed “This is an example of Anomaly Detection. In this case, the algorithm learns what is "normal" and what is "not normal", usually by observing the logs of the website”
- 64. What do youunderstand from Measures and Dimensions? Each field from the data source is automatically assigned a datatype (such as string, integer) and a role (dimension or measure) Aggregation applied on measures is ‘Sum’ by default but you can always change the default aggregation in the settings Can you tell what’s happening in the following cases? D. Predicting salary of an individual based his/her years of experience “This is an example of Regression. This problem can be mathematically defined as a function between independent (years of experience) and dependent variable (salary of an individual)”
- 65. Summary What is K-MeansElbow Method to choose K Clustering cars with K-means Classifying tumor with logisticWhat is logistic regression