Strict Standards: Declaration of JParameter::loadSetupFile() should be compatible with JRegistry::loadSetupFile() in /home/content/94/5783694/html/ on line 512

Warning: Cannot modify header information - headers already sent by (output started at /home/content/94/5783694/html/ in /home/content/94/5783694/html/ on line 107

Thanks For Your Feedback!

Five plus seven is? (answer as number)

Facebook Comments

Click to open/close fB Comments

Descriptive analytics

For a casino company, descriptive analytics could include pattern discovery methods such as customer segmentation, i.e., culling through a patron database to understand a patron’s preferred game of choice. Simple cluster segmentation models could divide customers into their preferred choice of games and this information can be given to the marketing department to create lists of baccarat players for a baccarat tournament. Market basket analysis, which utilizes association rules, would also be considered a descriptive analytics procedure. Casinos should use market basket analysis to bundle and offer promotions as well as gain insight into gaming, shopping and purchasing behavior. I will go into full detail on this in chapter three.

Pattern Discovery

Pattern recognition is a branch of machine learning that focuses on the recognition of patterns and regularities in data, although it is in some cases considered to be nearly synonymous with machine learning. Pattern recognition systems are in many cases trained from labeled "training" data (supervised learning), but when no labeled data are available other algorithms can be used to discover previously unknown patterns (unsupervised learning).

The terms pattern recognition, machine learning, data mining and knowledge discovery in databases (KDD) are hard to separate, as they largely overlap in their scope. Machine learning is the common term for supervised learning methods and originates from artificial intelligence, whereas KDD and data mining have a larger focus on unsupervised methods and stronger connection to business use. Pattern recognition has its origins in engineering, and the term is popular in the context of computer vision. 

In machine learning, pattern recognition is the assignment of a label to a given input value. An example of pattern recognition is classification, which attempts to assign each input value to one of a given set of classes (for example, determine whether a given email is "spam" or "non-spam"). However, pattern recognition is a more general problem that encompasses other types of output as well. Other examples are regression, which assigns a real-valued output to each input; sequence labeling, which assigns a class to each member of a sequence of values (for example, part of speech tagging, which assigns a part of speech to each word in an input sentence); and parsing, which assigns a parse tree to an input sentence, describing the syntactic structure of the sentence.

Pattern recognition algorithms generally aim to provide a reasonable answer for all possible inputs and to perform "most likely" matching of the inputs, taking into account their statistical variation. This is opposed to pattern matching algorithms, which look for exact matches in the input with pre-existing patterns. 

Pattern Discovery work that we have done in the past include:

  • Customer Segmentation – developed proprietary customer segmentation models that specifically identify a customer’s preference based off historical behaviour. These have been utilized in the gaming, wagering and retail industries and can be utilized for marketing, loyalty to a preference and strategic decisions. Views include, but are not limited to: brand, price, play day, play game, stake size, number of legs per bet.
  • Lifestage and Lifecycle Segmentation Models – in the retail industry, results of which were used to improve the marketing arm as well as understand the appropriate offering for a retail shopfront based on customer georgraphy profile.
  • Customer Database Segmentation Models – these include classical RFM models as well as risk and marketing models in the sports wagering industry.
  • Risk Management – developed an automated process to categorize whether individual sport wagers should be retained or bet back by a sports betting company. These models identified significant ROI both for bets being retained by the house, and those identified as not worth holding onto.
  • Data Reduction – coupled with a proprietary rating system, this information was used to derive a profitable staking plan based off the type of race run by a thoroughbred at its last start, specifically for the thoroughbred horseracing in Hong Kong.
  • Association Analysis – use by the Federal Health Department (Australia) to identify patient tracking through the healthcare system. Utitlised with provider information to understand whether corporate clinics were more likely to refer patients to their specialist services than private practitioners.