Understand Customer Needs for HANA Analytics – Analytics 3.0 – Webcast Summary
Understand Customer Needs related to SAP HANA Advanced Analytics: I watched this webcast today from SAP. I missed the first few minutes of the presentation.
Figure 1: Source: SAP
Figure 1 shows we’re towards the end of Analytics 2.0 “Big Data”
Now we are looking to Analytics 3.0 which provides “real impact to businesses”
Figure 2: Source: SAP
On Figure 2, the SAP presenter said there is a “talent shortage of data scientist” and a high learning curve
During this time, this tweet came in:
— Data Hacker (@sbashyal) February 13, 2014
This shows the job board for a data scientist from the Strata Conference this week.
A data scientist spends 60-80% of his/her time collecting data sources and less on analyzing the data.
Figure 3: Source: SAP
Figure 3 shows the 4 “pillars for decision processes”
As an example, a CRM person is on the phone with a customer, will get a lead score, to close a deal; this brings data close to decision making
This helps predict/act/optimize
“Bring computing to where the data is”
SAP has an open ecosystems – partnerships with SAS
Figure 4: Source: SAP
Figure 4 shows PAL algorithms which include native MPP (massive parallel processing) with classic data mining algorithms and are available where data lies
You can also connect to R packages
It includes unstructured data processing
Figure 5: Source: SAP
eBay uses HANA for data marts for their marketplace, showing the number of hits, what people are buying and what people are bidding.
Data needed to be “crunched”. This used to take 4-5 days
If there were changes to Google search engine, it took a day to find a problem and 4-5 days to figure out how to fix
After using HANA, same processing using time series and linear regressions went down to 3 hours; so they went from 4 to 5 days to 3 hours
The speaker suggested visiting http://www.saphana.com/community/about-hana/advanced-analytics
The idea incubator was the next topic covered. This is a crowd-sourcing program from SAP.
Figure 6: Source: SAP
Some customers are looking at social media based on stock price change and want the social media feedback
What social media will impact the stock the most?
Can a real-time prediction be made from stock change?
Figure 7: Source: SAP
There are two types of innovation as shown in Figure 8:
1) Open – like a hack day
2) Other site is closed innovation
Question & Answer
Q: What is the most popular HANA function for ideas?
A: Most popular idea is text analytics – social media and comments
Another one is predictive analysis
Q: Where can I learn about HANA
Q: I want to test an idea and not sure I am allowed to share data with strangers
A: three different ways – 1) hide customer data 2) simulated data 3) invited innovation – evaluate people before share
ASUG Annual Conference will have customer stories about Predictive with HANA