Downloaded 12 times
Uploaded byEntrepreneur / Startup
Machine Learning Algorithms in Enterprise Applications
Machine learning is being increasingly used in enterprise applications. Some key points: - Machine learning techniques like supervised learning, unsupervised learning and reinforcement learning are being used in applications for predictive analytics, descriptive analytics, and optimization. - Common machine learning algorithms like linear regression, logistic regression, SVM and neural networks are being applied to use cases in various industries for tasks like prediction, classification, clustering and recommendation. - Major companies like SAP, LinkedIn and DialogTech are using neural networks and other machine learning techniques for applications like spam detection, fraud detection, predictive analytics, customer segmentation and next best action recommendations. - Example use cases discussed are using machine learning for location-based marketing and promotions, monitoring
Machine Learning Algorithms in Enterprise Applications
- 1.
- 2.
- 3. ML in EnterpriseApps Image source: https://www.sap.com/products/machine-learning-foundation.htmlhttps://www.sap.com/products/machine-learning-foundation.html
- 4.
- 5. McKinsey Report Image source:https://www.youtube.com/watch?v=Op83nO706po
- 6. ML Terminology Image source:https://towardsdatascience.com/machine-learning-an-introduction-23b84d51e6d0
- 7. Process Involved Image source:https://towardsdatascience.com/machine-learning-an-introduction-23b84d51e6d0
- 8. ML Approaches ● SupervisedLearning(Predictive) ○ Learn mapping given dataset y(x) , D ={((xi,yi)}, e.g., MNIST classification ● Unsupervised Learning:(Descriptive) ○ Given only inputs , find interesting patterns D = {xi} e.g., Determine k cluster centers k ● Semi-supervised Learning ● Reinforcement Learning ○ How to act or behave when given occasional reward or punishment signals, e.g., how a robot learns to walk to a power outlet
- 9.
- 10. Linear Regression ● Astatistical model that attempts to show the relationship between two variables with a linear equation. ● Involves graphing a line over a set of data points that most closely fits the overall shape of the data. ● Shows the extent to which changes in a "dependent variable," which is put on the y-axis, can be attributed to changes in an "explanatory variable," which is placed on the x-axis. Image source: https://towardsdatascience.com/introduction-to-machine-learning- algorithms-linear-regression-14c4e325882a
- 11. Logistic Regression ● Methodfor analyzing a dataset ● There are one or more independent variables that determine an outcome. Image source: https://towardsdatascience.com/logistic-regression-b0af09cdb8ad
- 12.
- 13. SVM A Support VectorMachine (SVM) is a discriminative classifier formally defined by a separating hyperplane. Suppose you are given plot of two label classes on graph as shown in image (A). Can you decide a separating line for the classes? Given labeled training data (supervised learning), the algorithm outputs an optimal hyperplane which categorizes new examples Image A
- 14. Neural Networks Neural networksare a set of algorithms, modeled loosely after the human brain, that are designed to recognize patterns. Rnn: Networks that add connections feeding the hidden layers of the neural network back into themselves.
- 15. Application of LogisticRegression: ● Logistic regression is used when the response you want to predict/measure is categorical with two or more levels. Some examples are gender of a person , outcome of a football match ● Marketing: ○ A marketing consultant wants to predict if the subsidiary of his company will make profit, loss or just break even depending on the characteristic of the subsidiary operations. ● Human Resources: ○ The HR manager of a company wants to predict the absenteeism pattern of his employees based on their individual characteristic. ● Finance: ○ A bank wants to predict if his customers would default based on the previous transactions and history.
- 16. Application of LogisticRegression: ● Image Segmentation and Categorization ● Geographic Image Processing ● Handwriting recognition ● Healthcare : ○ Analyzing a group of over million people for myocardial infarction within a period of 10 years is an application area of logistic regression. ○ Prediction whether a person is depressed or not based on bag of words from the corpus seems to be conveniently solvable using logistic regression and SVM. ○ It is one of the best tools used by statisticians, researchers and data scientists in predictive analytics. ● It is one of the best tools used by statisticians, researchers and data scientists in predictive analytics.
- 17. ML in EnterpriseApplications ● Sales Recommendations and Predictions Example : Recommend related tickers, Predict next ticket from same customer ● Suggest products Supporting documentation to sales reps ● Build models Disparate sources of sales and marketing data ● Improve ROI
- 18. ML in VariousApplications
- 19. Neural Network -Applications 1. Finance: 2. Insurance: ○ Fraud detection, ○ Why an individual rejected their service 3. Operations management: ○ Optimize the functioning of equipment and extends its lifespan ○ Monitor the process, assist in optimization, detection of defective products 4. Retail: ○ Estimates which products were bought today, ○ How many times, and ○ What combination of products was bought 5. Marketing: ○ To arrange a productive target marketing campaign
- 20. Neural Network -Applications 6. Text Summarization: If a company wants to display key information from any literature within their apps or website, Text Summarization would be helpful. 7. Text Autofill or next text recommendation: Businesses looking to transform their data entry work by improving their workflow digitally can achieve faster automation 8. Language Translation Rather than hiring native translators to translate a massive volume of content, businesses can at least improve their translation process using Recurrent Neural Network 9. Call Center Analysis 10. Digital Asset Management in Marketing
- 21. How LinkedIn usesML algorithms ? ● LinkedIn uses neural networks along with linear text classifiers ○ to detect spam or ○ to detect abusive content in its feeds when it is created ● Use neural nets to help understand all kinds of content shared on LinkedIn ○ — ranging from news articles to jobs to online classes ○ — to build better recommendation and search products for members and customers. Source: https://www.cmswire.com/digital-experience/what-is-a-neural-network-and-how-are-businesses-using-it/
- 22. How DialogTech usesML algorithms ? ● DialogTech uses neural networks ○ to classify inbound calls into predetermined categories or ○ to assign a lead quality score to calls ● ML Actions performed based on the call transcriptions and the marketing channel or keyword that drove the call, For example, a caller who is speaking with a dental office may ask to ‘schedule an appointment.’ The neural network will seek, find and classify that phrase as a conversation, therefore providing marketers with valuable insights into the performance of marketing initiatives. Source: https://www.cmswire.com/digital-experience/what-is-a-neural-network-and-how-are-businesses-using-it/
- 23. Use Case 1:Customer Engagement and Commerce ● Able to design location-specific advertisements for specific products and distribute customized information to Facebook users. ● Consumers also receive location-relevant promotions at the right time on mobile devices. ● The promotional information displayed on mobile devices serves as shopping guidance in stores. ● The location-based marketing strategy through the use of social media has generated a sales uplift of 10% to 15%
- 24. Use Case 2:Hospital - Monitoring of Patient Care ● Gives a 360-degree view of patients, ● A fully integrated patient care lifecycle management solution, ● Covers all cases such as prevention, operation, recovery, and community or home care, ● The solution is mobile health app for patients and community doctors, ● Provides personalized online care plans on mobile devices issued to patients by hospital doctors, ● Integrate medical care provided by primary care physicians
- 25. Case Study :SAP Leonardo ML Foundation ● It provides an enterprise-grade platform for machine learning in the cloud.
- 26. SAP Applications ● SAPCash Application ○ Offers automation in finance, ○ Intelligent and Integrated Payment Clearing Automation for SAP S/4HANA powered by SAP Leonardo Machine Learning ● SAP Brand Impact ○ Automatically analyzes large volumes of videos, ○ Video Analytics to Measure Brand Exposure Faster, Accurately, and at Scale ● SAP Service Ticket Intelligence ○ Automatically categorizes customer tickets and proposes solutions
- 27.