AI for our Rock Paper Scissors Lizard Spock game in Java on the SSaurel's Channel

NextMove.java

private Item nextMove(Item prev) {
  if (nbThrows < 1) {
    // first move => we can't use Markov Chain prediction
    // so we use a random on the Item list
    return Item.values()[RANDOM.nextInt(Item.values().length)];
  }

  // we try to predict next Item choosen by the user by reading data in our Markov Chain
  // for the prev entry in the array
  int nextIndex = 0;

  for (int i = 0; i < Item.values().length; i++) {
    int prevIndex = prev.ordinal();

    if (markovChain[prevIndex][i] > markovChain[prevIndex][nextIndex]) {
      nextIndex = i;
    }
  }

  // Item probably played by the user is in nextIndex
  Item predictedNext = Item.values()[nextIndex];

  // we choose amongst item for which this probably item loses
  List<Item> losesTo = predictedNext.losesTo;
  return losesTo.get(RANDOM.nextInt(losesTo.size()));
}