The project displays on how to use clustering methods to help study Environmental Science when playing the strategic video games. The main focus is on Phase II where in the game, soldiers attack nearby weaker opponents. The objective of the game is to create kingdoms, produce resources (wood, stone) to build castles and then form an alliance to defeat enemies. Therefore, the main question is “how to build a powerful troop/castle”? First, we need to know the types of troops needed to build a castle. In phase II, there are up to 40 types of troop units which are mostly u sed for attacking and defending. There are seven characteristics for each troop unit which are Melee/Range Defense, Melee/Range Attack, Travel Speed, Looting Capacity, and Food Consumption.
The objective of this paper is to introduce an advanced multivariate statistical methodology for advancement through the Empire 4 Kingdoms strategic video game. A novel Troop Power Index (PI) is derived that captures the objective of developing a powerful troop army and expanding the size of one’s own kingdom. The Power Index applies specified weighting coefficients that consider how to build a powerful troop army which can both attack and defend well and have good looting capability and speedy travel capacity. After the PI was established, the following were considered (1) summation/ weighting of the untransformed PI, (2) Z-Transformed PI, and (3) Non-Parametric transformed PI, in order to reduce the variance bias and the impact to outliers among the seven troop characteristics. (1) through (3) were compared to determine their unique characteristics in the distribution of relative PI ranks among the troops. Next, supervised clustering based on power index (4 clusters) and unsupervised JMP hierarchical clustering were used (8 clusters) to select the top vital few troops from each cluster based on PI ranking. The Non-Parametric method was chosen to build the transfer function and to optimize the troop units since most of the troop characteristics were highly skewed. The troop types considered fit to the Kingdom Expansion Strategy were reduced from 45 to 15 based on the Non-parametric Troop PI Ranking. In future work, Transfer Function Sensitivity Analysis may be explored within the Troop Constraints in order to further optimally select the troop units to meet the current Kingdom expansion strategy.
0 Comments
This paper would introduce a novel educational methodology of playing a strategic video game-Empire Four Kingdoms. In modern Big Data World, Artificial Intelligence algorithm is powerful for engineering problem solving to discover the complicated science or/and mechanics. Separating Statistics from Math can draw practical decision and conduct risk assessment. Several modern Artificial Intelligence algorithms are adopted to build a statistical model of optimizing the troop design and understanding the Military Science. Environmental Science and Natural Resources (Stone, Wood, and Food) are also addressed to educate the importance of protecting our world against Technology Advancing.
|