ezeetester

Quick Tip: Anomaly Detection to build trust with Customers

In Automation, Bills, Invoices, Machine Learning on November 21, 2016 at 4:05 pm

In a recent incident, one of my friends took his old vehicle for servicing.  Among other things, a small bulb was replaced and the bill for just that part was a whopping Rs 300/-.  The explanation given to him was that the bulb costs Rs.150/- and the labor was Rs. 150/-. Whereas the same would cost nearly half the price a few months ago. Courtesy a history of maintaining invoices, he was able to question the prices.

He was also at the receiving end for some of his other bills, when he saw some extra charges which were never charged in his earlier bills (which he had maintained).  It was later found out, that these were some “errors”.   I wonder how many people would really do all this.

For people who do not store so many bills/invoices and keep track of unusual deviations in them.  They could be ripped off.

Solution: In these days of digital records (including bills and invoices), a machine learning algorithm called Anomaly Detection could help.  A company could basically spot the outliers or anomalies and strive to eliminate them thereby building trust with their customers.  For the customers, they could benefit if certain flags are reported in their bills whenever a large variation in prices is seen in their invoices for many billing products.

Some bills, payslips etc, could contain breakups of various categories, and any missing category or additional categories can also be flagged.

All in all, a win win for both Customers and Businesses

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: