There’s a popular television game show called “Wheel of Fortune” that features enthusiastic contestants solving word puzzles by guessing one alphabet letter at a time. Each word puzzle is classified under a specific category name that gives the players a further clue to uncovering the round’s hidden phrase.
A popular category in the game, called Before & After*, consists of two phrases (or names) that are combined by a word that ends the first phrase and starts the second phrase. For example, “washing machine” & “machine learning” = “washing machine learning.”
This linguistic mash-up is actually pretty fascinating when you consider the countless processes, approaches and intricate methods employed in the analytical makeup of today’s sophisticated revenue management systems – all combined through analytics. From data modeling to analytical market segmentation to accounting for the effects of price elasticity, the analytics under the hood of your RMS accomplish a vast amount of extremely difficult (and different!) tasks. For example, you may have caught this recent blog post explaining how advanced analytical technology sorts and organizes your data – comparable to doing laundry in a washing machine.
In today’s wide realm of revenue technology and analytics, there’s been a “new” algorithmic process called machine learning garnering a lot of air time in recent revenue management discussions. Interesting enough, machine learning is actually not new to the hospitality industry at all – advanced revenue management technology has been incorporating this process, where appropriate, into their analytics for quite some time.
SAS® Institute, IDeaS’ parent company, defines machine learning as “a method of data analysis that automates analytical model building.” Basically, this process enables computers to find hidden insights without being explicitly programmed where to look (think: web searches or targeted marketing advertising.) IDeaS’ advanced revenue management systems have long been using the process of machine learning, in conjunction with statistical methods, to produce cutting-edge forecasting and decision optimization.
However, machine learning doesn’t come without its complexities: the learning methods it needs to consider, its relationship with both data mining and statistics, and the types of problems its application works well for. If you’re interested in learning even more about the detailed process of machine learning, download our “What’s Old is New Again: Machine Learning in Revenue Management Technology” white paper.
The methodical make-up of revenue technology’s analytical engine is extremely intriguing; peeling back its intricately-woven layers of approaches, processes and methods is an additional evaluation metric in today’s powerful technology – especially when your hotel is looking to spin the wheel for maximum revenue performance.
*The name for this category varies throughout different international regions. For example, in Germany, it is referred to as “2 in 1” and in the UK, it is called “The Common Word.”