Someone in the world develops dementia every 3 seconds! Around 50 million people in 2017 already have dementia. The gloomy forecast: reaching 75 million in 2030 and over 131 million in 2050. The total estimated worldwide cost of dementia is USD billion in 2015, which represents 1.09% of global GDP. By 2018, the global cost of dementia will rise above a US$ trillion (The global Voice on Dementia). Around 75% of the dementia people have Alzheimer. These figures explain why lot of efforts and means are invested to diagnose AH as early and professionally as possible. The research is focused on two main objectives. The first is "Prediction": Early AH detection by predicted-based algorithms, for those who currently have or probably will have AH. The second is "Efficiency": Improving the current diagnostics process; specifically, regarding the time-consuming parameter. All these, by computers' data mining algorithms: mainly clustering, classification and neural-networks; by using advanced computers' architectures, such HADOOP, and by other computer-based means.