ALPHAMINER 2.0
用户手册
哈工大-香港大学商务智能联合实验室
2006 年 6 月
目录
CHAPTER 1 ALPHAMINER系统 .................................................................................................1
1.1 开始 .........................................................................................................................................1
1.1.1
菜单与工具条
.................................................................................................................4
1.1.2
案例管理窗口
.................................................................................................................6
1.1.3
选择和配置工作区
.........................................................................................................8
CHAPTER 2 数据访问.....................................................................................................................1
2.1 从文件中输入数据 .................................................................................................................1
2.2 从数据库中输入数据 .............................................................................................................4
CHAPTER 3 数据探索.....................................................................................................................1
3.1 数据探索 .................................................................................................................................1
3.1.1
柱状图
.............................................................................................................................2
3.1.2
多变量绘图
.....................................................................................................................5
CHAPTER 4 数据转换.....................................................................................................................1
4.1 设置属性 .................................................................................................................................2
4.2 增加属性表达式 .....................................................................................................................5
4.3 异常值处理 .............................................................................................................................6
4.4 缺失值 .....................................................................................................................................7
4.5 标准化 .....................................................................................................................................8
4.6 抽样 .........................................................................................................................................9
4.7 二值化 ...................................................................................................................................11
4.8 数据选择 ...............................................................................................................................13
4.9 类别属性转换 .......................................................................................................................16
4.10 数值转换 ...............................................................................................................................18
4.11 数据集可处理化 ...................................................................................................................20
CHAPTER 5 建模.............................................................................................................................1
5.1 关联规则 .................................................................................................................................2
5.2 KMEANS ..................................................................................................................................4
5.3 决策树 .....................................................................................................................................6
5.4 指数回归 .................................................................................................................................9
5.5 朴素贝叶斯 ...........................................................................................................................11
5.6 线性回归 ...............................................................................................................................12
5.7 多层感知器 ...........................................................................................................................13
5.8 RBF网络................................................................................................................................15
5.9 序贯最优化算法 ...................................................................................................................16
5.10 WKMEANS.............................................................................................................................19
CHAPTER 6 评估.............................................................................................................................1
6.1 评估.........................................................................................................................................2
6.1.1
混淆矩阵
.........................................................................................................................2
6.1.2
评价图
.............................................................................................................................4
CHAPTER 7 部署.............................................................................................................................1
7.1 部署 .........................................................................................................................................1
AlphaMiner2.0 · 用户手册 · 哈工大-香港大学商务智能联合实验室 (2006). i
表格目录
表
1-1
案例菜单
......................................................................................................................................5
表
1-2
窗口菜单
......................................................................................................................................5
表
1-3
高级菜单
......................................................................................................................................5
表
1-4
工具条按钮
..................................................................................................................................6
表
2-1
数据访问模块
..............................................................................................................................1
表
2-2
数据访问模块的连接限制
..........................................................................................................1
表
3-1
数据探索模块
..............................................................................................................................1
表
3-2
数据探索模块的连接限制
..........................................................................................................1
表
3-3
数据探索
–
散布图的定制参数
.................................................................................................6
表
4-1
数据转换模块
..............................................................................................................................1
表
4-2
数据转换模块的连接限制
..........................................................................................................1
表
4-3
事务化过程生成的属性
............................................................................................................21
表
5-1
建模模块
......................................................................................................................................1
表
5-2
建模模块的连接限制
.................................................................................................................2
表
5-3
关联规则
–
参数
........................................................................................................................2
表
6-1
评估模块
......................................................................................................................................1
表
6-2
评估模块的连接限制
..................................................................................................................1
表
6-3
评估模块的参数设置
..................................................................................................................2
表
6-4
评估
–
实际结果与预测结果
.....................................................................................................4
表
6-5
评估
–
混淆矩阵中另外的性能参数
.........................................................................................4
表
6-6
评估
–
评价表的参数
..................................................................................................................5
表
6-7
评估
–
评价图的说明
.................................................................................................................8
表
7-1
部署模块
......................................................................................................................................1
表
7-2
部署模块的连接限制
..................................................................................................................1
图形目录
图
1-1 ALPHAMINER
系统的图形界面
....................................................................................................2
图
1-2
知识库中的数据挖掘案例
..........................................................................................................2
图
1-3
案例信息
......................................................................................................................................3
图
1-4
案例示图面板中的数据挖掘案例
..............................................................................................3
图
1-5
鼠标右键弹出该模块的下拉菜单
.............................................................................................4
图
1-6
建模结果
......................................................................................................................................4
图
1-7
案例管理窗口
..............................................................................................................................6
图
1-8
案例分组列表框
..........................................................................................................................7
图
1-9
查看案例
......................................................................................................................................7
图
1-10
编辑案例
....................................................................................................................................8
图
1-12
增加工作区
................................................................................................................................9
图
1-15
切换工作区
..............................................................................................................................10
图
2-1
从文件中输入数据
–
拖放“从文件中输入数据”模块
...........................................................2
图
2-2
从文件中输入数据
–
设置参数
.................................................................................................2
T
图
2-3
从文件中输入数据
–
选择一个
EXCEL
工作表
........................................................................2
图
2-4
从文件中输入数据
–
调入数据
...................................................................................................3
图
2-5
从文件中输入数据
–
成功执行
...................................................................................................3
图
2-6
从文件中输入数据
–
查看元数据
...............................................................................................3
图
2-7
从文件中输入数据
–
查看数据
...................................................................................................4
图
2-8
从数据库中输入数据
–
选择数据库驱动
.................................................................................4
图
2-9
从数据库中输入数据
–
连接数据库服务器
...............................................................................5
图
2-10
从数据库中输入数据
–
根据表名选取数据
.............................................................................5
图
2-11
从数据库中输入数据
–
用
SQL
查询选取数据
..........................................................................6
AlphaMiner2.0 · 用户手册 · 哈工大-香港大学商务智能联合实验室 (2006). ii
图
2-12
从数据库中输入数据
–
保存设置
.............................................................................................6
图
2-13
从数据库中输入数据
–
执行该模块
.........................................................................................7
图
2-14
从数据库中输入数据
–
查看元数据
.........................................................................................7
图
2-15
从数据库中输入数据
–
查看数据
.............................................................................................8
图
2-16
从数据库中输入数据
–
导出数据
.............................................................................................8
图
2-17
从数据库中输入数据
–
将数据保存到
EXCEL
文件中
...............................................................9
图
3-1
数据探索–
执行该模块
............................................................................................................1
图
3-2
数据探索
–
数据集的基本信息
.................................................................................................2
图
3-3
数据探索
–
属性列表
.................................................................................................................3
图
3-4
数据探索
–
属性统计信息
.........................................................................................................3
图
3-5
数据探索
–
单个属性的分布
...................................................................................................4
图
3-6
数据探索
–
交叉属性的分布
...................................................................................................5
图
3-7
数据探索
–
可视化所有的交叉属性分布
...............................................................................5
图
3-8
数据探索
–
多变量绘图
...........................................................................................................6
图
3-9
数据探索
–
选择散布图的
X
轴与
Y
轴
........................................................................................7
图
3-10
数据探索
–
为每个绘图选择颜色属性
.................................................................................8
图
3-11
数据探索
–
数据实例细节信息
..............................................................................................8
图
3-12
数据探索
–
用矩形选择数据实例
..............................................................................................9
图
3-13
数据探索
–
用多边形选择数据实例
.....................................................................................9
图
3-14
数据探索
–
在散布图中绘制折线
.......................................................................................10
图
3-15
数据探索
–
利用折线选择实例
.............................................................................................10
图
4-1
设置属性
–
属性细节
...............................................................................................................2
图
4-2
设置属性
–
将属性类型由数值型转化为类别型
...................................................................3
图
4-3
设置属性
–
将属性类型由类别型转化为数值型
...................................................................3
图
4-4
设置属性
–
设置目标属性
.........................................................................................................4
图
4-5
设置属性
–
去除一个属性
.......................................................................................................4
图
4-6
设置属性
–
使用一个属性
.........................................................................................................5
图
4-7
增加属性表达式
–
创建一个新的空属性
.................................................................................6
图
4-8
增加属性表达式
–
创建一个由数学表达式赋值的新属性
.....................................................6
图
4-9
异常值处理
–
替换异常值
.........................................................................................................7
图
4-10
缺失值
–
替换缺失值
...............................................................................................................7
图
4-11
标准化
–
标准正态分布
...........................................................................................................8
图
4-12
标准化
–
线性规范化
...............................................................................................................9
图
4-13
抽样
–
随机采样
..................................................................................................................10
图
4-14
抽样
–
1-IN-N.........................................................................................................................10
图
4-15
抽样
–
前
N
个
........................................................................................................................11
图
4-16
二值化
–
选择类别
................................................................................................................12
图
4-17
二值化
–
选择类别
................................................................................................................12
图
4-18
二值化
–
二值化后创建的新属性
.........................................................................................13
图
4-19
二值化
–
二值化后的数据值
.................................................................................................13
图
4-20
选择
–
根据范围选取
.............................................................................................................14
图
4-21
选择
–
根据数值型属性进行选取
.........................................................................................15
图
4-22
类别属性转换
–
自动转化为数值
.........................................................................................16
图
4-23
类别属性转换
–
映射为其它类别
.......................................................................................17
图
4-24
类别属性转换
–
映射为数值
.................................................................................................17
图
4-25
数值转换
–
自动转化为类别值
.............................................................................................18
图
4-26
数值转换
–
根据区间的数目进行离散化
.............................................................................19
图
4-27
数值转换
–
根据区间的宽度进行离散化
.............................................................................19
图
4-28
数值转换
–
为
AGE
属性创建的类别属性
...............................................................................20
图
4-29
数值转换
–
将
AGE
属性值映射到类别中
.............................................................................20
图
4-30
数聚集可处理化
–
转化前的例子数据
..................................................................................21
图
4-31
数聚集可处理化
–
事务化后的元数据
.................................................................................22
图
4-32
数聚集可处理化
–
事务化后的数据
.......................................................................................22
AlphaMiner2.0 · 用户手册 · 哈工大-香港大学商务智能联合实验室 (2006). iii
评论0
最新资源