Social Business Intelligence Framework. Copyright 2012 Deloitte Development LLC. All rights reserved.

Transcription

1 Social Business Intelligence Framework

2 Key Insight / Takeaways Business Outcomes Insightful Brand Analysis Improved Customer Experience Benchmarked Performance Revelation of Market Trends/Opportunities Sentiment Analysis Technology Enabler Extraction of data from the Social Networking sites Classification of Data Integration of Social data with Business data Presentation of the derived analytics Key Features Reusable, Scalable & Configurable Industry Agnostic Demo ready based on real data Interactive & Intuitive Dashboards 2 Sentiment Analysis POC

3 Framework Extraction - Extract the data from Social Networking sites Analysis & Classification - Cleansed & Classify unstructured data through algorithm Presentation - Map social data with business parameters Extraction Analysis & Classification Presentation Cross Industry Framework: Can be tailored for any Industry 3 Sentiment Analysis POC C&IP E&R FSI LSHC PS TMT

4 Methodology Extraction Analysis & Classification Presentation Extraction: Retrieve data from social media websites Improve Data Quality: Remove hyperlinks, spam, commercials, etc. Categorize data : Group data on features like, customer experience, product feedback, etc., Cleansing: Remove noise words and Translate Internet Slang into authentic English Prepare The Training Data: Prepare training set to educate the machine learning algorithm Train The Algorithm: Use classified messages to educate the Machine Learning Algorithm on how to classify messages Classification: Feed extracted cleansed messages to trained algorithm for classification Standardization: Convert the classified data into the structure form Design Dashboard: Identify dimensions/kpis to visualize the data. Define dashboard template/reports Consume Data: Design and Implement interface to consume the data Develop: Develop reusable widgets, dashboards and reports to visualize the KPIs of interest 4 Sentiment Analysis POC Needs some customization for each industry/sector

5 Framework Illustration Extraction Analysis & Classification Presentation Shopping at A*B is something I love to do My bad shopping experience at a A*B last night is now up on the site Had the worst cashier at A *B today! CUSTOMER SERVICE HOT NEW PET... earn 3% # cash back with # Shop At Home coupon codes! Check here! Shop Mission Home Exclusively at A*B! ADAPTER SPAM FILTER DATABASE Training the Algorithm Shopping at A*B store is something I love to do Assigning Sentiment to the Tweet Had the worst cashier at A*B today! My bad shopping experience at a A*B last night is now up on the site

6 Demo 6 Sentiment Analysis POC

7 Interactive Dashboards Sentiment Breakup Top 5 influencer s Sentiments Sentiment by time for every click on the overall sentiment Overall sentiment of the Users selected in the left panel Mash up View Brand Selection User Interactivity Animated Drill Down Capability Heat Map Overall sentiment breakup by Brand Brand Sentiments across states Sentiment comparison by time Sentiment comparison by features of the brands Drill Down: Post count for each brand for a selected state 7 Sentiment Analysis POC

8 Intuitive Dashboards Market opportunities across states. Color coded to visualize the sentiment Drill Down on map to view posts and features breakup Brand Selection Mash up View Brand Selection User Interactivity Animated Drill Down Capability Heat Map Sentiments by time View Comments 8 Sentiment Analysis POC

9 Roadmap Current State Framework defined and implemented to explore intelligence from social media Extracts, cleanses and classifies user comments on different brands and products from one of the social networking websites Implemented and fine-tuned Naïve Bayes Algorithm to derive user sentiments written in English Provides a visualization interface to discover and identify predefined KPI s and analyze user sentiments Next Steps Call Center Notes Analysis - Validate if Model can be expanded for call center note analysis Analysis & Classification Include additional Algorithm to improve accuracy for multiple entities Compatibility with mobile devices - Will build the dashboards compatible with tablets & other mobile devices Integrate Social data with existing data warehouse Will come up the methodology Integrate POC with Social CRM Work with SAP team to integrate with CRM 9 Sentiment Analysis POC

10 10 Sentiment Analysis POC

11 For more details Sandeep Sharma [email protected] Sriram Nagarajan [email protected] Vadiraj Muthya [email protected] 11 Sentiment Analysis POC

12 About Deloitte Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited, a UK private company limited by guarantee, and its network of member firms, each of which is a legally separate and independent entity. Please see for a detailed description of the legal structure of Deloitte Touche Tohmatsu Limited and its member firms. Please see for a detailed description of the legal structure of Deloitte LLP and its subsidiaries. Certain services may not be available to attest clients under the rules and regulations of public accounting. This publication contains general information only and is based on the experiences and research of Deloitte practitioners. Deloitte is not, by means of this publication, rendering business, financial, investment, or other professional advice or services. This publication is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional advisor. Deloitte, its affiliates, and related entities shall not be responsible for any loss sustained by any person who relies on this publication.