Machine Learning Session by Artivatic AI Data Labs
4 likes56,163 views
The document outlines the requirements and algorithms for effective artificial intelligence (AI) and machine learning (ML), focusing on classification and clustering techniques. It discusses various ML algorithms such as supervised and unsupervised learning, including details on classification algorithms like naive Bayes and neural networks, as well as clustering methods like k-means and DBSCAN. Additionally, it emphasizes areas of application such as fraud detection, healthcare, market segmentation, and more.
Machine Learning Session by Artivatic AI Data Labs
- 1. AI/ML Workshop By Artivatic Data Labs Team In Association with 21 Jan 2017 (By Ananth G, Omkar & Layak) This confidential document is legal property of Artivatic Data Labs Private Limited.
- 2. Content Requirements For Making Good AI ML Algorithms Classification Algorithms Clustering Algorithms Advance ML Algorithms Reference This confidential document is legal property of Artivatic Data Labs Private Limited.
- 3. Requirements For Making Good AI This confidential document is legal property of Artivatic Data Labs Private Limited. Data Preparation AI Algorithm and Model Creation Autonomous and Interactive Learning Scalability
- 4. Machine Learning Algorithms • Categories of ML algorithms • Classification / Supervised Learning • Clustering / Unsupervised Learning • Supervised algorithms can apply what has been learned in the past to new data. • Unsupervised algorithms can draw inferences from datasets. This confidential document is legal property of Artivatic Data Labs Private Limited.
- 5. Classification Classifies data into given set of predefined categories. Supervised Accuracy depends on the training data set. Training Dataset New Data Class of new dataThis confidential document is legal property of Artivatic Data Labs Private Limited. Classification Algorithm Classification Rule
- 6. Where To Use Classification Fraud Detection Machine Vision Natural-Language Processing(NLP) Bioinformatics Healthcare and Disease Prediction This confidential document is legal property of Artivatic Data Labs Private Limited.
- 7. Classification Algorithms Naive Bayes • Use Bayes’ theorem, • Where, A and B are events and P(B) ≠ 0 P(A) and P(B) are the probabilities of observing A and B without regard to each other. P(A | B), a conditional probability, is the probability of observing event A given that B is true. P(B | A) is the probability of observing event B given that A is true This confidential document is legal property of Artivatic Data Labs Private Limited.
- 8. • Neural Networks • Based on the structure and functions of biological neural networks. • Every link between two neurons are weighted • Theoretically we can design any level of the Neural Network This confidential document is legal property of Artivatic Data Labs Private Limited.
- 9. This confidential document is legal property of Artivatic Data Labs Private Limited.
- 10. • Some Common Classification Algorithms • Nearest Neighbour Algorithms • Support-Vector Machines • Decision Tree Classifier • Hierarchical Classifier • Random Forest This confidential document is legal property of Artivatic Data Labs Private Limited.
- 11. Clustering Creating clusters depends on the similarity between given data. Unsupervised Data within a cluster are more similar than the data between different clusters. This confidential document is legal property of Artivatic Data Labs Private Limited.
- 12. Where To Use Clustering Analysis of Human Personality Market segment Finance Sector Analysis City Planning Global Climate This confidential document is legal property of Artivatic Data Labs Private Limited.
- 13. Clustering Algorithm • K-means clustering • Create K clusters depending on distance between individual data points • Distance between intra-cluster points are lower than distance between inter-cluster points • Easy to Implement • High Time Complexity This confidential document is legal property of Artivatic Data Labs Private Limited.
- 14. K-mean Algorithm This confidential document is legal property of Artivatic Data Labs Private Limited.
- 15. • DBSCAN • Density-Based Spatial Clustering of Applications with Noise • Capable of ignoring Noise This confidential document is legal property of Artivatic Data Labs Private Limited.
- 16. Advance Machine Learning Algorithm Artificial Neural Networks Deep Learning Genetic Algorithms Competitive Learning This confidential document is legal property of Artivatic Data Labs Private Limited.
- 17. Any Questions? This confidential document is legal property of Artivatic Data Labs Private Limited.
- 18. Thank You. Write to us: [email protected] Visit Us: Artivatic Data Labs Private Limited 25 & 26, 2nd Floor, AVS Compound, 80 Feet Road, Koramangala 4th Block, Bangalore -34 Website: www.Artivatic.com