Machine Learning in Venue

How does AI assist to a venue marketing?

There are 3 requirements for using machine learning in business, and I’d like to describe 3 situations in venue marketing field as below.

1. The venue marketing problem can be formulated as a machine learning problem.

I’ve working on venue management field over 19 years, so I’m very interested in solving visitor management problem in a venue. I’d like to solve the issue of ‘predicting’, which is about personalized events recommendation based on the analysis of personal visit pattern. I think machine learning can predict human behaviors based on the analysis of its input data. It’s not the chaos but possible solution because human actions usually have specific reason on specific situation. Venues like convention centers or museums don’t provide any curated events list but just send out a bunch of events information to their annual visitors. Of course, visitors never read carefully about these pile of events list, moreover venues don’t know who are their loyal advocate or negative anti-fans. If venues can offer personalized events list including exhibition, convention or events, then venues service quality would be enhanced and these give positive impact to venue brand in terms of competitive strategy of differentiation. We’ve already experienced the success case through Netflix and Amazon’s personalized product recommendation system.

2. There should be enough relevant data available that could be used by machine learning algorithms.

There are abundant relevant data for analyzing and sensing of visitors’ pattern. Convention centers or museums have many visitor information regarding who they are, why they visit, what they see, how they come, when they want to visit and what they eat during the visit time, etc. About 10 million people visit convention center in Seoul, and they provide personal information for their visiting every year. However, the data set is originally not from venues but from each event organizers, so that it is not easy getting visitor information. Especially Korea prohibit the leakage of personal information by law. If event organizers and venues can share their input data and analyze visitors pattern, it would be benefit for two side regarding venue branding and service offerings.

3. The system has enough regularity in it that there are patterns that can be learned.

MICE events including convention, tradeshow, event are usually opened annually. So every year similar visitors participate regularly these events for their own reason. It is good for machine learning because it shows continuous pattern for their visiting, so that machines can learn pattern of visitors every year. Venues can get the appropriate input data in terms of regular base, and therefore machine learning can produce well-designed output information for venue’s visitor marketing.

The more experience machine learning algorithms get, the better they will become. It will be an extraordinary future a few years later as the technology continues to mature.