Image

How Machine Learning Will Impact
Event Management

By :Niraj Shah
CEO


Forward-thinking businesses have their eyes on machine learning. Google has long emphasized its priorities in pursuing machine learning and artificial intelligence. Of course, they're not the only ones. Twitter’s recent acquiring of Magic Pony Technology indicates their keen interest in machine learning as well.

Apart from being a news item and a buzzword, machine learning and its developments have great implications on business and broader society. Specifically for the events industry, what can we look forward to, in terms of how machine learning will impact events?

In this article, we will explore current applications of machine learning and make thoughtful predictions of how machine learning can be applied within events management. But first, let’s lay down some key definitions.


What’s the Difference: Machine Learning, A.I. and Beyond

In machine learning, the computer is able to automatically improve itself through experience. It that sense, it’s self-learning. Based on data that’s fed to or collected by the computer, the computer is able to make new decisions by itself to improve on efficiency and achieve its goals.

It’s a part of the broad umbrella of A.I. or artificial intelligence, the quest to design computers that mimic yet exceed human capacities, and provide practical solutions to real-world problems.

Data mining, another buzzword, is the process of collecting and understanding vast amounts of digital information. Because computers act on information, data mining is the foundation for A.I. and machine learning.

Algorithms are problem-solving formulas. They are the rules or directives that a computer follows when processing information and completing operations.

Deep learning, one more trending concept, takes machine learning to the next level. In deep learning, the machine is designed to function like human neural networks and be capable of higher-level abilities, such as understanding language and recognizing objects.  

The above concepts are inter-related. In the media, these terms are defined broadly and sometimes used interchangeably. To distill these concepts to an essence, we can say that computers are becoming smarter and more capable. The implication is that we can take advantage of all this for greater productivity.

Current Practical Applications of Machine Learning

Machine learning is often used to digest and analyze huge volumes of information, which humans cannot quickly and effectively process by themselves. The result is many real-world applications that we already experience everyday.

Some of the most common applications of machine learning include:


  • Netflix and Amazon recommendations that “learn” your preferences
  • Web search results that adapt to your viewing history and demographic
  • Real-time web and mobile ads that “seek” to be relevant to you
  • Email spam filtering
  • Text-based sentiment analysis
  • Prediction of equipment failure
  • Demand forecasting
  • Price optimization
  • Fraud detection
  • Image and pattern recognition
  • Self-driving cars
  • Robotics

As you can see, we use machine learning across many industries and it’s already a part of our daily lives. When it comes to events planning, we can say that machine learning has some peripheral applications already. For example, it affects how our event pages rank in search results. But we want to imagine the future of machine learning in this industry. Based on cases of current machine learning applications, we can predict future uses and benefits for the events industry.

How Machine Learning Can Benefit Events Management


1. Inform your planning and decision-making

One of the most popular applications of machine learning right now is for prediction and decision-making. Computers can read and analyze much more information than human minds can. Relying on computation for decision-making gives businesses a greater competitive edge.

Take professional sports, for example, and the analytics company Second Spectrum. Second Spectrum collects and analyzes data on athletes’ performance and uses this data to help coaches and teams to win. Half of NBA teams now rely on Second Spectrum software to manage and coach teams.

In addition to strategizing in sports, machine learning can assist in making hiring and recruiting decisions. In an experiment by McKinsey & Company, three algorithms were used to predict which job candidates out of a pool of 10,000 applicants were most suited for a certain company. The forecasts produced by computation closely matched real-world hiring results, demonstrating the effectiveness of machine learning to analyze résumé data and make reliable recommendations.

Moreover, in the above experiment, the computer selected slightly more female applicants than real-world hiring managers, suggesting that machine learning can further help us overcome preconceptions and make unbiased decisions.

Brian Uzzi, a Kellogg School of Management professor echoes this idea when he says, “Through human-machine partnerships, leaders will be able to strip away latent biases and make more empirical decisions, leading to more creative and insightful decisions.”

In terms of events management, machine learning can assist in the selection of panel experts and speakers. The use of machine learning for the process of speaker selection could possibly help some industries and events to overcome a tendency towards male-only panels. It can also save event planners vast amounts of time in comparing speaker qualifications and work histories, if they could rely on data analysis software.

Similarly, machine learning could be employed in the process of venue hunting. In the same way that Netflix suggests relevant movies and Amazon finds suitable products, a venue-search platform could help event planners source the right venue for their event, and suggest several appropriate options based on organizers’ history and preferences.


2. Allow you to make faster, more accurate decisions

Hotel managers are already using machine learning to assist in a key aspect of their business: price optimization. Algorithms can determine optimal room price at any given moment by analyzing real-time supply and demand and looking at competition from surrounding hotels, vacation homes and related types of accommodation. Calculations that take hours for humans to compute (after which, the results would be outdated), an algorithm can determine in real time and adjust prices automatically.

Beyond this, hotels know that different types of customers are happy to pay different prices for the same room. So algorithms are employed to predict what cost a highly satisfied customer would be willing to pay for a return visit, versus what cost a price-sensitive customer would be wiling to pay. In this way, price optimization is achieved not only in reflection to current supply, demand, and competition, but also optimized to the individual buyer, based on the buyer’s purchasing history and shopping behavior.

The kind of price optimization currently used in hotel management could be carried over to events management as well. Large events often have different price tiers and a variety of program and accommodation packages. They offer early bird pricing, flash sales, group pricing and various promotions. How can event managers know the optimal price to set each ticket type? What is the best timeframe for an early bird or flash sale? Algorithms can help decide these details, and predict when there will be a rush of buyers (the time to hold off on discounts) or when there will be a lull in sales (the time to bring on the marketing push). This sort of predictive analysis can optimize the profitability of events, just as it is optimizing the income of hotels.


3. Predict sentiment and preference of attendees

Sentiment prediction is another popular application of machine learning. Also called opinion mining, sentiment analysis helps organizations to determine public opinion on a certain product or topic, based on analysis of text -- usually web comments, social media posts and online conversations. The Obama administration, for example, used sentiment analysis to determine public opinion towards policy announcements during the 2012 election.

In short, sentiment analysis tells you how your audience feels, what they like or dislike. In this way, sentiment analysis is useful for event planners who want to know what artists to book for next year’s event, what speakers to invite, and what themes to choose for a future conference. Will your future conference participants want to hear about AI applications or UX design? How popular and impactful were last year’s talks and workshops, and which ones should you repeat or improve upon? Sentiment analysis can help event planners gain insight into areas such as these, giving planners predictive wisdom for events down the road.

In addition to analyzing sentiment, machine learning can also cater to attendee preferences. Music platform Spotify, for example, is able to adapt to listeners’ music preferences and suggest appropriate tracks. One day, an events-sourcing platform may be able to achieve the same level of personalization. Algorithms could be used to suggest relevant workshops and seminars for conference attendees, helping to cut down on attendees’ time in browsing and researching events. Event-goers could be alerted to upcoming concerts in their area for bands and musical genres matching their tastes. Event merchandise and conference resources could be suggested purchases and add-ons automatically offered to event page visitors as well.


4. Curate and categorize conference content

Machine learning allows us to curate, tag and categorize vast amounts of content. This library of content can be accessed by viewers seeking information on various topics, and algorithms can help viewers find what they’re looking for.

It’s exactly what YouTube has been doing, using algorithms to suggest related content and help viewers find the most relevant query results. YouTube’s algorithms imitate Google’s deep learning algorithms used for search recommendations. YouTube’s VP of product management Johanna Wright tells TechCrunch, “We want YouTube to communicate this feeling that it understands you.”

This artificial feeling of understanding and personalization achieved by machine learning can be carried over to events management. For organizers who have a history of running high numbers of events, imagine if all your past workshop, seminar and presentation content were to be curated and tagged. Attendees, members, or paid subscribers could then access this content library, where they could receive help from recommendation algorithms in discovering interesting playlists and relevant content.


5. Achieve greater customer segmentation

Data allows us to discover more about our customers, giving us a better understanding of how best to connect with them. In one famous case, retailer Target figured out that one of its customers was pregnant even before her own family knew. Based on what Target had analyzed about this customer’s buying history, the company proactively mailed pregnancy and baby product coupons in hopes of winning her as a lifelong customer. It greatly surprised her family to be tipped off by the store.

Extreme cases aside, data analysis can help businesses and event planners suss out new customers. For example, you may be able to segment who on your mailing list is new in their industry and interested in career development events, or who is based in New York and possibly able to attend your next conference. Machine learning can make this segmentation process automatic. A potential customer would be categorized and put on a relevant marketing campaign without any intervention from you or a team member, saving you time and resources.


6. Take advantage of coming A.I. service and hospitality trends

In an outlier example of machine learning applied to the hospitality industry, Japanese hotel Henn na is staffed by robots. Upon entry, guests are greeted by three multilingual machines: a woman humanoid, a dinosaur bot, and a futuristic bot. A robotic arm sits in the foyer and places guests’ belongings into cabinets for safekeeping -- a robotic cloakroom.

The hotel’s founder Hideo Sawada created Henn na in an effort to save labor costs and to highlight A.I. innovation. Sawada says, “In the future, we’d like to have more than 90 percent of hotel services operated by robots.”

If Henn na hotel is a sign of the future, and if more hotels and businesses incrementally adopt robotic assistance, A.I. will flow into events experiences as well. By and large, events are run in hotels and other hospitality establishments. When hotels are machine-enhanced, so too will be events. One day, you may no longer have to send a staff or volunteer to replenish the room’s coffee supply or to adjust audio-visual settings. These small details may be automatically monitored and managed by smart machines.


At Eventgrid, we’re passionate about making your events a success. Whether you’re running a concert or a conference, we supply the tools to make your event fly. From online registration to mobile-friendly event pages, we’ve got you covered. Talk to us today for a free Eventgrid demo.




A lot goes into building a successful event. We’re here to help. 

Request a demo today