Request a demo We can arrange a demo to show you the capabilities of our products

Our products are used to:

_reduce the uncertainty of Machine Learning models,

_prioritize enterprise data efforts,

_support experts in the ML loop,

_improve the quality of ML models, especially in multi-class settings with complex ontologies,

_reduce data footprint and compactify ML models so as to be used by the Internet of Things applications,

_improve gaming experience via more challenging and realistic AI in games,

_create intelligent advisory systems from pre-compiled building blocks.

Sensei Demonstrative Realization: Tactical Troops Analytics

Tactical Troops Analytics (TTA) is a portal for players of the game Tactical Troops that provides both personalized and global analytics. It also provides processed data for the game developer and publisher.

See below what elements of Sensei are demonstrated in TTA:

  • Log processing and data ingestion module.
    Sensei comes with various ways to obtain data e.g. querying public API or interfacing with the game. Tactical Troops pushes the data asynchronously to one of the Sensei database models.
    TTA uses Django and PostgreSQL, which is one of the possible solutions.
  • Aggregated global and personalized statistics.

Global statistics include elements such as players' ranking, wins percentages, kills per game, deaths per game, parameters of using various weapons, gadgets etc. Personalized statistics give more insight into a particular player's favourite choices (e.g. favourite weapon, map or loadout) as well as their proficiency in exploiting them to their advantage.

  • Rich Replay Viewer
    It enables not only to analyze your past games in a slightly simplified tactical view but also contains advanced features such as heatmaps, important events on the timeline and game changing moments.
    10 types of heatmaps are available, e.g. places visited the most by players, common sites of deaths or where players position their units at the end of the turns.
  • Player Style

In addition to the personalized statistics, players are profiled using various unique style traits.
Sensei helps to normalize, store and visualize those traits. The visualization is based on a polygon (aka spider's web). The most interesting features, which are detected using Sensei tools, are shown by default.

  • Players Clustering
    Sensei enables to take any features of players and use them to perform clustering.
    Similar players belong to the same cluster. The visualization is based on the UMAP algorithm.
    It gives various insights e.g. how many similar groups there are, where I am as a player, how strong the players in my cluster are, how players in various clusters differ from me etc.
    It allows to see how the playing style affects the ranking of players. For instance, are the top players in the same cluster?
  • Formations
    Formation analytics offers a few kinds of information:
    • how players group their units in squads
    • how large the squads are
    • which types of units go well together
    • what are the typical distances between units on a given map
  • Popular loadouts
    Loadouts in Tactical Troops consist in choosing units and equipment before each match.
    There is some kind of analogy of a loadout to a "build" or a "deck" known from other games.
    Loadouts can be aggregated, abstracted with various levels of granulation thanks to Sensei mechanisms.
    On the website, we present a matrix of win rates between the most popular loadouts defined in two levels of granulation.
  • Loadout Recommendation
    Sensei guesses players' preferences by analyzing their playstyle and can recommend loadouts to players that are suitable for them.
    Players can set custom constraints: e.g. they want a particular unit to appear in a loadout. Sensei will complement their preferred starting choices.
  • Winning odds predictor
    Predictive analysis of how is more likely to win a match can be used in various ways.
    We can use the winning odds to introduce a betting mechanism.
    Players can also see how the system evaluates their chances based on various parameters e.g. the game mode or map to be played.
    It can be used for example to introduce a handicap or to prepare for a certain opponent. After all, it is fun to win against the odds.
  • Player Growth
    Sensei provides the visualization of the player's ranking change in time.
    More significant changes are detected and the system automatically finds the most suitable reason that made the player gain or lose the ranking.
    Such a reason can, for instance, be the change in the favourite (most used) loadout.

        Make Sensei your mentor!

        Let’s set up a meeting!

        Send