diff data_mining_theory/4_16.md @ 0:1b00d07674d7

add note
author Takahiro SHIMIZU <anatofuz@cr.ie.u-ryukyu.ac.jp>
date Thu, 18 Apr 2019 16:40:43 +0900
parents
children
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/data_mining_theory/4_16.md	Thu Apr 18 16:40:43 2019 +0900
@@ -0,0 +1,51 @@
+- We can turn data into knowledge
+
+
+- Superviser Algorithm use Labels, detail corpus
+
+- regression
+    - 回帰
+- classification
+
+- Preprocession
+    - 準備
+    - transforming the data
+- Learning
+- Evaluation
+    - use test data
+- Prediction
+    - use unknown data
+
+# WhyPython?
+
+- Numpy
+- Pandas
+- Scikit learn
+
+# QA
+
+## What's feature many learning costs?
+
+- Could give us example test regression task?
+    - surrary?
+
+# Chap2
+
+## Perceptron Algorith
+- Classfication algorithm that cased on MCP neuron model
+    - vectors or input and weight
+    - use binary data
+- w .. wight
+- x .. vector data
+
+# calculate
+
+z = `\sum{w_i*x_i}`
+
+
+# 英語
+- regression
+- classification
+- increasing [増加する]
+    - His artworks are likely to increase in value.
+    - The number of people over the age of 65 increased to nrearly 23 millions in 2011.