使用Python撰寫程式實現資料挖掘

Abstract

資料挖掘為對極龐大資料的探詢，探詢當中蘊含的規則、模式的方法。而會探究這樣的方法便是因為現今我們能夠蒐集到的資訊是極為龐大的，而想透過人腦直接利用這龐大的資料，讓人一眼便看出當中的規律在數據龐大之下是幾乎不可實現的。而相對的，電腦則是處理龐大數據的行家，因此才會想研究資料挖掘在電腦上的實現方法。 而在現今便有著許多方法被討論出來，其中我們想實現的一是Apriori先驗演算法，其方法能讓我們快速的找出資料當中的frequient patterns和其association rules，二則為DNN模型，透過訓練模型，試著對資料進行預測分類。 在Python上循著這兩種方法實作出可執行的程式後，我們最後可以透過這兩者從資料內尋找到一定的規則。對於找到的規則，有助於我們更瞭解資料間的關係，從中推斷出有利的資訊從而利用之。

Data mining is a method of discovering patterns in large data sets.We explore such a method because the information we can collect today is extremely large, and if you want to directly use this huge data through the human brain. It is almost impossible under the huge amount of data.On the other hand, computers are experts at handling huge amounts of data, so we want to study the implementation of data mining on computers. Nowadays, many methods have been discussed. Two of the methods we want to implement are Apriori algorithm and DNN model. Apriori algorithm, which allows us to quickly find the frequent patterns and its association rules in the data sets. DNN model. Train the model and try to predict and classify the data. After we follow these two methods to implement executable programs on Python, we can finally find certain rules from the data through these two methods. The rules found will help us better understand the relationship between the data, infer beneficial information from it and use it.