ML-aaS for Advertisers

 

Advanced Machine Learning Algorithms

to deterministically detect invalid traffic. 

AI Driven

State of the art Generalised Discriminant Analysis(GDA), Network Analysis and Graph Theory and other advanced machine learning algorithms. 

Abnormal Behaviour Detection

First of the kind, fraud coefficients namely Alpha, Beta & Gamma makes our reports interpretation easy and fast. 

Independent

Technology would help risk professionals audit their data assets without any conflict of interest. 

Cyber Technology

Cyber Fraud as a problem cannot be solved by technology itself, it needs a cyber technology intervention. 

SINGLE TEST TO DETERMINISTICALLY DETECT 

PRESENCE OF FRAUDULENT ADVERTISING TRAFFIC

Auto Identify Anomalous Consumer Behaviour/Clusters  

1. Evaluate behaviour weekly/monthly
 

2. Highlight anomalies
 

3. Check your traffic for quality

Deterministically Differentiate Anomalous Behaviour 

  1. Self visualise bot traffic, engaged and disengaged users. 
     

  2. Study the fraud clusters in size and strategise to reduce cluster size
     

  3. Save weekly reports for reference

Highlight and Identify Source 

  1. Self identify bot traffic sub/sources.
     

  2. Self visualise fraud traffic reduction on source takedown. 

DETERMINISTIC TEST FOR ORGANIC HIJACKING

Traditional Method

Com Olho Method

Studying time to install or landing behaviours is a good way to estimate amount of organic traffic that is being hijacked. This allows the advertiser to understand if they are at a risk of financial and performance losses. This method is a widely adopted method among marketers. 

Studying time to install or landing behaviours is indeed a good way, but because of its simplicity, it can be easily programatically manipulated making the invalid traffic valid.  

DETERMINISTIC TEST FOR BOT MIXING

Traditional Method

The industry is deploying fraud teams to look into wide variety of data points, in order to highlight abnormality or finding organised data structures in order to understand fraud. For bot mixing, there isn’t any method available to segregate good and bad traffic. 

 



 

Some of the manual research efforts done : 

 

  1. Device Analysis

  2. IP Analysis

  3. Geographical knowledge.

  4. Duplication Analysis

Com Olho Method

Bot traffic generated using device farms, SDK Spoofing, APK drops is often delivered with almost human mimicking attribution data. We deploy state of the art patented technology to distinguish valid and invalid traffic.

Uncover digital fraud plaguing your business using Nex-Gen Tech

contact@comolho.com | +91 810 500 5490

  • Black LinkedIn Icon
  • Black Twitter Icon

Com Olho is a leading cloud-based machine learning as a service platform. We make it easy for CXO's to understand their consumers using behaviour analytics, machine learning and data science. 

e : contact@comolho.com

© 2019-2020 COM OLHO IT PRIVATE LIMITED

  • Facebook
  • Grey LinkedIn Icon

Made with      in India