Professional Documents
Culture Documents
Clustering: k-means
Classification
Association
Propensity modelling
Apriori-apprioori
Association algorithm: Put simply, the apriori principle states that if an itemset is
infrequent, then all its subsets must also be infrequent
Both X and Y can be placed on the same shelf, so that buyers of one item would
be prompted to buy the other.
Promotional discounts could be applied to just one out of the two items.
Advertisements on X could be targeted at buyers who purchase Y.
X and Y could be combined into a new product, such as having Y in flavors of X.
While we may know that certain items are frequently bought together, the question is,
how do we uncover these associations?
Besides increasing sales profits, association rules can also be used in other fields. In
medical diagnosis for instance, understanding which symptoms tend to co-morbid can
help to improve patient care and medicine prescription.
Definition
Association rules analysis is a technique to uncover how items are associated to each
other. There are three common ways to measure association.
Measure 1: Support. This says how popular an itemset is, as measured by the
proportion of transactions in which an itemset appears. In Table 1 below, the support of
{apple} is 4 out of 8, or 50%. Itemsets can also contain multiple items. For instance, the
support of {apple, beer, rice} is 2 out of 8, or 25%.
Table 1. Example Transactions
If you discover that sales of items beyond a certain proportion tend to have a significant
impact on your profits, you might consider using that proportion as your support
threshold. You may then identify itemsets with support values above this threshold as
significant itemsets.
Measure 2: Confidence. This says how likely item Y is purchased when item X is
purchased, expressed as {X -> Y}. This is measured by the proportion of transactions
with item X, in which item Y also appears. In Table 1, the confidence of {apple -> beer}
is 3 out of 4, or 75%.
One drawback of the confidence measure is that it might misrepresent the importance
of an association. This is because it only accounts for how popular apples are, but not
beers. If beers are also very popular in general, there will be a higher chance that a
transaction containing apples will also contain beers, thus inflating the confidence
measure. To account for the base popularity of both constituent items, we use a third
measure called lift.
Measure 3: Lift. This says how likely item Y is purchased when item X is purchased,
while controlling for how popular item Y is. In Table 1, the lift of {apple -> beer} is
1,which implies no association between items. A lift value greater than 1 means that
item Y is likely to be bought if item X is bought, while a value less than 1 means that
item Y is unlikely to be bought if item X is bought.
BFL is a library that delivers a set of commonly used Financial Algorithms,
To be able to utilize the BFL algorithms you need to install the AFL (Application
Function Library) that matches the revision number of your HANA box. AFL will deliver
BFL plus PAL (Predictive Analysis Library)
FYI: the SAP ECC default/initial settings are: alpha = 0.2; beta =
0.1; delta = 0.3 and gamma = 0.3
IOT: Intel Retail Sensor(Intel RSP) Collects data from sensors in the
store and RFIDS from Products---Then RSP feeds into SAP Dynamic
Edge Processing Server --- SAP Dynamic Edge Processing Server
Monitors store data for critical events and dervies insights based on
business context for store to take action---Insync with
Remote Synchronize is useful for large datasets for trends but less
useful for real time actionable analytics at store(which is supported by
light weight SAP products closer to the store---Cloud
1) Transaction Availabilty for Remote sites(TARS) which allows ERP
transactions even when transactions to the IT Data center is
slow
2) SAP SQL anywhere for Server,desktop,mobile and remote office
3) SAP Streaming Lite(SDS)
4) SAP HANA Remote Sync
1) Superior Customer Service
2) Detect Item shrink quickly and more accurately then cycle
counting
3) Drive real time requests to restock shelves
4) Optimize check out times
Machine Learning:
The newest frontier in machine learning involves demand forecasting and
related functions. Introduced three years ago, demand forecasting that
uses machine learning can link assortment, space, price and fulfillment into
a single plan, factoring in time of year, weather and information on
competitive products, sell through, customer traffic and demographics.
Demand forecasting allows online and offline retailers to generate more
precise forecasts than traditional time series approaches. This yields more
accurate inventory levels in stores, online and warehouses. In-stock
positioning improves, there are fewer markdowns and better ROIs.
Toolsgroup;
Most demand planning software can factor in seasonalitylike the
fact that more ice cream is sold in summer than in winter. But
sometimes seasonality can become so extreme or complex that it is
not as well suited to normal regression analysis-based techniques.
For instance, at one customer machine learning sifts through SKU-
Locations to identify clusters of products with similar
seasonality profiles, recognizing more than 200 micro-climates
and their seasonal timing variations.
Social Sensing:
Weather and Macroeconomics
Supply the transaction and click the Display button and all the table access by the
particular transaction code will appear.
I'm working as a JR. ABAPer & frequently I encounter this situation when I'm given a
report either with fields from a structure or no fields at all. I have following doubts:-
1) Who's responsibility it is to find out the fields complete with table name?
Functional consultant or ABAPER?
2) Is there some std. way of finding out these fields?
1) First, goto SE11, and enter your structure name. Goto the field that holds your data,
and double-click on the element name. Once inside the element, do a where-used list
for that element, searching tables only. Then go into each table, and see if you can
find the one holding your data. There may sometimes be a huge number of tables
displayed, but a lot of them will be empty.
2) Get the name of the program behind your transaction, goto SE80, and enter the
program name. 99% of the time it will be part of a module-pool and bring up the pool.
Goto the dictionary structures, and search each table there for the one holding your
data.
3) Open a new session with transaction ST05, select SQL Trace, click on the Trace
On button, and go back to your transaction (while leaving the session with ST05
open). Submit your transaction, and go back to the ST05 session. Click on the Trace
Off button, then select the List Trace button. Continue with standard selections, and a
BASIC TRACE LIST will appear. search thru the tables displayed under
ObjectName.
In ECC 6.0, you can ge the similar functionality through a function module.
Check the Package - SEST, Function Group - SEA1 and Function Module -
RS_PROGRAM_TABLES.
This FM provides the same result. Just Provide Object Type as 'T' (in case of Transaction code,
likewise P in case of Program), Object Name as 'XD01' (Transaction Code for Customer Master) and
execute. It will provide all the tables which is used by this Transaction.