現今是高度追求數位化以及智慧化的時代，全球的產業結構、消費方式等正因各種新興技術崛起而有巨大改變，應用新技術的產品、產業、服務等如雨後春筍一般出現，並且不斷的創新、發展、擴大應用範圍，讓現在的產業經營模式持續轉型，各產業期望能夠應用人工智慧來優化產業供應鏈，配合大數據分析，讓企業能提前預測並快速掌握客戶的需求，並配合智慧製造在交易時以更有效率的方式提供快速的服務以及優良的產品品質，帶給客戶良好的交易體驗，讓企業獲取最大利益。
本論文的目的即是研究如何將傳統業務交易過程的資訊及數據雲端化並系統化，並利用機器學習和大數據分析，結合網頁前端技術、資料庫管理去設計及實現一個智慧型的業務系統，並研究在此架構下如何實現減少業務時間成本、提高工作效率及成交率、智慧化分析拜訪模式等商業願景，並配合智慧製造，把人工智慧應用在整體產業供應鏈。在面對不同的客戶族群時，為了讓企業能快速因應各客戶族群客製化地制訂經營策略及營業目標，必須讓取得客戶需求到產品生產的過程更智慧化。本研究探究如何在業務系統引入機器學習及大數據分析技術，整合業務從客戶端獲得的資訊和數據，應用分析這些資訊以提升企業的決策智慧。

Nowadays, we live in a period that people highly seek digitalization and artificial intelligence. The Industrial structure and consumption patterns in the world is hugely changing because of various new technologies developing. The industries, products, and services applied new skills widely appear in a short time. The new skills are continuously innovating, developing, and expanding their application area, making the existing industry business models constantly transforming. Every industry expects that artificial intelligence would optimize the supply chain and big data analysis would make enterprises able to forecasting and quickly grasping the clients’ demands. Then the enterprise could affiliate smart manufacturing to offer faster service and better products when trading in a more efficient way. Therefore, they can provide great trading experience for clients, and obtain great profit.
The purpose of this thesis is to study how to Cloudization and Systematization the data of a traditional transaction, applying machine learning and big data technology, combined with the website front-end technology and the database management system to design and to implement a smart business system, and under such a framework how to implement business expects like reducing time cost, increasing work efficiency and turnover rate, and analyzing visiting modes smartly, and then affiliate smart manufacturing to apply artificial intelligence to all the parts of supply chain. When trading with various group of clients, the process of getting the client’s demand until producing products must be smarter in order to make the enterprise quickly set the business strategy and goal for every client group. This Research explores how to introduce machine learning and big data technologies into a business system, and further integrates the data from clients and analyze the data to promote the decision-making wisdom of the enterprise.

摘 要 i
ABSTRACT iii
誌 謝 v
目 錄 vi
表 目 錄 viii
圖 目 錄 ix
一、導 論 1
1.1研究背景 1
1.2研究動機 1
1.3研究目的 2
1.4研究架構 2
二、相 關 文 獻 3
2.1機器學習簡介 3
2.2機器學習的流程 3
2.3機器學習類別 4
2.4特徵工程 5
2.5機器學習演算法 5
2.5.1 Logistic Regression 5
2.5.2 決策樹 6
2.5.3 KNN 6
2.5.4 K-means聚類算法 7
三、系統架構分析與設計 8
3.1系統分析 8
3.2系統架構 10
四、系 統 實 作 11
4.1實作環境 11
4.2系統開發工具介紹 11
4.3業務系統 13
4.3.1客戶資料管理 14
4.3.2客戶詳細資料 16
4.3.3撰寫週/日報表 20
4.3.4業務審核管理 25
4.4機器學習應用 27
五、結 論 29
參 考 文 獻 30

參 考 文 獻
[1] 林大貴，TensorFlow + Keras 深度學習人工智慧實務應用，博碩，2017。
[2] 王正林，數據科學家成長之路 Python 大戰機器學習 數據科學家的第一個小目標，電子工業，2017。
[3] 劉立民，Python機器學習(第二版)，博碩，2018。
[4] “Apache Mahout,” Apache Software Foundation, [線上]. Available: https://mahout.apache.org/developers/developer-resources.html. [Retrieved: 8 2019].
[5] “MariaDB,” [線上]. Available: https://mariadb.org. [Retrieved：8 2019].
[6] “Bringing MySQL to the web,” [線上]. Available: https://www.phpmyadmin.net/. [Retrieved：8 2019].
[7] “CodeIgniter User Guide — CodeIgniter 3.1.10 documentation,”[線上]. Available: https://www.codeigniter.com/userguide3/index.html. [Retrieved: 8 2019].
[8] 奚江華, 網頁程式設計ASP.NET MVC 5. x 範例完美演繹(附範例程式), 磬峰, 2018。
[9] “Anaconda Distribution” [線上].
Available: https://www.anaconda.com/distribution/. [Retrieved: 8 2019].
[10] “Some information about the Jupyter Project and Community” [線上]. Available: https://jupyter.org/about. [Retrieved: 8 2019].