How Artificial Intelligence could benefit the MMO industry? I believe machine learning can help build more stable social dynamics that are so essential to MMOs.
Guilds and groups are a defining feature of MMOs. They help new players learning the ropes and stay engaged, they facilitate sharing of resources and the creation of stable groups, necessary to beat the hardest dungeons. They constitute an essential part of the player experience. For a new player, finding good team mates can make the difference between sticking to the game and quiting. One of the conclusion of a quantitative study on World of Warcraft is:
Overall our data gives the impression that growing a guild to a significant size and sustaining it over time is a difficult task, perhaps more so than managing groups in other contexts. From a game producer’s standpoint this is clearly problematic since, as we have seen earlier, guild membership encourages players to play more and to group more.
Using the qualitative research I’ve done on MMOs, I identified two pain points related to this problem. Both could significantly benefit from machine learning algorithms.
1st pain point: finding a guild you like
When you’re a new player, or just someone whose guild has become inactive, the process to find a (new) guild is messy. There is no clear path. You’re lurking on world chat and in the forums, sometimes ask questions. You might join a number of different guilds, with different characters, struggling to build lasting relationships. Working through trial and errors, it often takes a long time to find a guild that is a good fit.
2nd pain point: recruiting new members
Online communities are notoriously difficult to maintain, and it takes a lot of hard work to constantly recruit new members to maintain a guild at a functional level. Guild turn-over rates are very high. Guild leaders and officers spend a lot of time recruiting and managing the group using different tools and strategies. The burden of guild management can sometimes overcome the most motivated players. The departure of a leader can be at the origin of the disparition of a guild. Other common causes are the natural attrition rate and internal conflicts between members.
Machine learning contribution
Machine learning is perfectly suited to virtual worlds because data are extensively available on players and their characters: play time and duration, playstyle, level, abilities, stats, etc.
Based on this data, a virtual neural network could be trained to make the best match between existing guild members and other players. Leaders would get a list of potential recruits, and players looking for a guild a list of groups that best fit their level and playstyle.
It would save a lot of recruiting time for leaders and officers, and a lot of guess work for new players. If guild members are better suited to one another, this would likely decrease the rate of attrition and turn-over, making for more stable, and therefore more engaging guilds and gameplay.