Social Media AnalyticsPowered by Data Science
Presented by Navin Manaswi
Flow of the Presentation● Social Media : What is it? How big is it? What are its types? How much important is it
for businesses? Use cases
● Big Data Analytics : What is it? How much important is it for businesses? How do we do it ? Use cases Success Stories Opportunities across globe
● Power of Data Science in Social Media (Big Data) Analytics : How can you leverage for your business? Powerful insights Sentiment Analysis Social Network Analysis Top Influencers Challenges
Global Digital Snapshot UAE Digital Snapshot
Social Media Data has
VolumeVelocityVeracityVariety
We need to use Big Data and Data Science to make use of it
Importance of Social Media in Industries Nowadays people tend to depend on the advice of friends and known people while making important decision related to any product and service. And, they are using social media in the form of social networking, social shopping and social bookmarking more than ever as a source to be able to make important decisions wisely
Visa, Wells Fargo, AMEX and JPM Chase try to move ahead in their quest for dominance, and so, the competition for the top slot is getting intense day by day
If you track these major banks on social media and analyze the buzz around them, you will come to know the INSIGHTS about these banks and their products and services. For the deep dive analysis, we focus on three key factors –
1. Share of Volume,
2. Sentiment Score and
3. Top Topics of Discussion about these brands
Social Media in FinanceUse case
Glimpse of Social Media Analytics :
Sentiment Analysis, Opinion Mining, Social Network Analysis,
Wordcloud, Top Influencer
What is Social Media Data ?Any data available on social media which can be leveraged to get actionable insightsExample of social media data:
SharesLikesMentionsImpressionsHashtag usageURL clicksKeyword analysisNew followersComments
How does Social Media Data help ?Once social media data is collected, it is measured or analyzed to get actionable insights for Digital Marketing Manager, Brand Manager, Digital Marketing Strategist, Social Media Manager, Event Manager and Product Manager
“Social media data acts as the ingredients to your meal and Social media analysis acts as your recipe”
Social Media Types
Social Media Types
Social Networking
Microblogging
Social News
Media Sharing
Blog Comments and Forums
Social Bookmarking
Advantages of Big DataKey areas where Big Data can help in Marketing are:
- Create customer segments based on huge data of transaction and other attributes
- Implement more targeted marketing campaign for specific geographies or individual customers
- Create upsell and cross-sell strategies based on transactional behaviours
- Identify which promotion strategy will yield the best results in a specific chain or cluster of stores
- Determine which new product options are the most profitable or the least risky to pursue
- Better assess product price elasticity before implementing price changes
Enable Micro- Market Campaign management
Send personalised Marketing messages to consumers based on algorithmix personalized recommendation engine so as to achieve high conversion ratio.
Optimize Promotions
Increase merchandising effectiveness by leveraging social sentiment insights across geographies over a period of time
Forecasting real time demands
Forecast demands by using machine learning algorithms more accurately than ever
Improve on-shelf performance and reduce out of stocksImprove retail store performance and inventory turnsImprove demand planning and reduce wastages.
Improve campaign target segment responseIncrease sales and market shareImprove customer loyalty and brand affinity
Improve customer segmentationUnderstand customers’ need betterIncrease upsell dramaticallyIncrease cross-sell dramatically
Big Data Analytics Use cases Business Outcome
Social Media Big Data : Analytics Process
Collection of Data from various sources
Extraction and Storage of Data
Data Preparation and Data Analytics
Data Visualization : Dashboards and Reports
Data Science, Machine Learning,Natural Language Processing
Hadoop Clusters,Hive, Pig on top.
Interactive Visualization, BI Tools
APIs, Flume
Advantage: If you get the insights, you can expect 10 times higher chance of clicking the ad when the ad is shown So you can achieve 10 times higher revenue
There are more than 200,000,000 Facebook users with college degrees, and they have been each served 100 ads.let's say that Facebook wants to know which ads work best for people with college degrees. Let's say there are 200,000,000 Facebook users with college degrees, and they have been each served 100 ads
That's 20,000,000,000 events of interest, and each "event" (an ad being served) contains several data points (features) about the ad: what was the ad for? Did it have a picture in it? Was there a man or woman in the ad? How big was the ad? What was the most prominent color? Let's say for each ad there are 50 "features"
This means you have 1,000,000,000,000 (one trillion) pieces of data to sort through. If each "piece" of data was only 100 bytes, you'd have about 93 GB of data to parse. That's pretty big (but still arguably not quite into "big data" territory), but you get the idea
Why is Big Data Analytics very important ?
Aim : To maximize click ads
Insights: Which features of ad are most effective in getting college grads to click ads ?
Big Data Analytics AchievementGoogle famously showed that they could predict flu outbreaks based upon when and where people were searching for flu-related terms :
Big Data Opportunity in World
Big Data Architecture
Hadoop : OverviewA scalable fault-tolerant grid system for data storage and processing
• Commodity hardware
• HDFS: Fault-tolerant high bandwidth clustered storage
• MapReduce: Distributed data processing
• Works with structured and unstructured data
Hadoop Design Principles
System shall manage and heal itself
• Automatically and transparently route around failure
• Speculatively execute redundant tasks if certain nodes are found to be slow
• Performance shall scale linearly
• Compute should move to data
• Simple core, modular and extensible
Power of Data Science in Social Media Analytics1. Sentiment Analysis
2. Social Network Analysis
3. Identification of Top Influencers
4. Identification of most related words
5. Understanding the main concerns of customers
6. Tracking public sentiments real time
7. Identification of social network
8. Tracking sentiments for rival products/services
Social Media Analytics : For Legoland
Sample of Social Media Analytics
Social Network Analysis : For Legoland
Social Media Analytics : For LegolandSample of Social Media Analytics
Take Away Points
1. What is Social Media?
2. Relevance of Social Media in Industries
3. What is Big Data ?
4. Social Media Big Data : What, How and Why ?
5. Data Science and Social Media Analytics
6. Use cases of Social Media Analytics
Thank You!
Ready for your Questions