跟讀練習: Gaussian Naive Bayes, Clearly Explained!!! - 通過影片學習英語口語
選擇練習模式
每種模式都會開啟全屏練習室。
107 句
如果句子過短或過長，請點選 Edit 進行調整。
1
Beep boop boop beep boop boop beep boop.
2
StatQuest.
3
Hello, I'm Josh Starmer and welcome to StatQuest.
4
Today we're going to talk about Gaussian Naive Bays, and it's going to be clearly explained.
5
Note.
6
This stat quest assumes that you are already familiar with the main ideas behind multinomial Naive Bays.
7
If not, check out the quest!
8
The link is in the description below.
9
This StatQuest also assumes that you are familiar with the log function, the normal or Gaussian distribution,
10
and the difference between probability and likelihood.
11
If not, check out the quests.
12
The links are in the description below.
13
Imagine we wanted to predict if someone would love the 1990 movie Troll 2 or not.
14
So we collected data from people that love Troll 2, and from people that do not love Troll 2.
15
We measured the amount of popcorn they ate each day.
16
How much soda pop they drank.
17
and how much candy they ate.
18
The mean for popcorn for the people who love Troll 2 is 24.
19
and the standard deviation is 4.
20
And a Gaussian, or normal distribution, with mean equals 24 and standard deviation equals 4 looks like this.
21
Likewise, the average amount of popcorn for people who do not love Troll 2 is 4.
22
and the standard deviation is 2.
23
And that corresponds to this Gaussian, or normal distribution.
24
Now we calculate the mean and standard deviation for Soda Pop for people that love Troll 2.
25
and draw the corresponding Gaussian distribution.
26
Then we do the same thing for the people that do not love Troll 2.
27
Lastly, we draw the Gaussian distributions for candy.
28
Gaussian Naive Bayes is named after the Gaussian distributions that represent the data in the training dataset.
29
Now someone new shows up.
30
and says they eat 20 grams of popcorn.
31
and drink 500 milliliters of soda pop, and eat 25 grams of candy every day.
32
Let's use Gaussian Naive Bayes to decide if they love Troll 2 or not.
33
The first thing we do is make an initial guess that they love Troll 2.
34
This guess can be any probability that we want, but a common guess is estimated from the training data.
35
For example, since 8 of the 16 people in the training data loved Troll 2, the initial guess will be 0 .5.
36
So we'll put that up here so we don't forget.
37
Likewise, the initial guess for Does Not Love Troll 2 is 0 .5.
38
So let's put that here so we don't forget.
39
Oh no, it's the dreaded terminology alert.
40
The initial guesses are called prior probabilities.
41
Now, the score for Love's Troll 2 is...
42
The initial guess that the person loves Troll 2, times the likelihood that they eat 10 grams of popcorn given that they love Troll 2.
43
Note: the likelihood is the y -axis coordinate on the curve that corresponds to the x -axis coordinate.
44
and we multiply that by the likelihood that they drink 500 milliliters of soda pop given that they love Troll 2.
45
times the likelihood that they eat 25 grams of candy given that they love Troll 2.
46
The initial guess that someone loves Troll 2 is 0 .5.
47
The likelihood for popcorn is 0 .06.
48
The likelihood for soda pop is 0 .004, And the likelihood for candy is...
49
A really, really small number.
50
When we get really, really small numbers, it's a good idea to take the log of everything to prevent something called underflow.
51
The general idea of Underflow is:
52
Every computer has a limit to how close a number can
53
get to zero before it can no longer accurately keep track of that number.
54
When a number gets smaller than that limit, we run into underflow problems and errors occur.
55
So we use the log function to avoid underflow.
56
Any log will do, but the natural log, or log base E, is the most commonly used log in statistics and machine learning.
57
So we take the log of everything, and the log turns the multiplication into the sum of the individual logs.
58
The log base E of 0 .5 is...
59
-0 .69 The log of 0 .06 is negative 2 .8.
60
The log of 0 .004 is -5 .5.
61
And the log of this really, really small number is negative 115.
62
Now we just add this up.
63
And we get negative 124.
64
So the log of the Love's Troll 2 score is negative 124.
65
BAM!
66
Now let's calculate the score for Not Loving Troll 2.
67
We start with the initial guess that someone does not love Troll 2.
68
times the likelihood that they eat 20 grams of popcorn given that they do not love Troll 2.
69
times the likelihood that they drink 500 milliliters of soda pop, times the likelihood that they eat 25 grams of candy.
70
So let's plug in the numbers.
71
Beep, boop, beep.
72
Boop.
73
and take the log of everything.
74
and that turns the multiplication into the sum of logs.
75
Now we just do the math.
76
Beep, boop, boop, boop.
77
and we get -48.
78
And since the score for Does Not Love Troll 2 is greater than the score for Loves Troll 2,
79
We will classify this person as someone who does not love Troll 2.
80
Double bam!
81
Note: When we look at the raw data, it almost looks like we should have classified this person as someone who loves Troll 2.
82
After all, they ate a lot more popcorn than the average person who doesn't love Troll 2.
83
and they drank as much soda as the average person who loves Troll 2.
84
However, the big thing is that they ate a lot more candy than the people who loved Troll 2.
85
and the log of the likelihoods for candy are way different.
86
And this difference is what made us classify the new person as someone who does not love Troll 2.
87
In other words, Candy can have a much larger say in whether
88
or not someone loves Troll 2 than popcorn and soda pop.
89
And this means we might only need candy to make classifications.
90
We can use cross -validation to help us decide which things, popcorn, soda pop, and /or candy, help us make the best classifications.
91
Shameless self -promotion.
92
If you don't already know about cross -validation, check out the quest.
93
The link is in the description below.
94
Triple bam.
95
Oh no, it's another shameless self -promotion.
96
One awesome way to support StatQuest is to purchase the Gaussia Naive Bayes StatQuest Study Guide.
97
It has everything you need to study for an exam or job interview.
98
It's seven pages of total awesomeness.
99
And while you're there, check out the other StatQuest study guides.
100
There's something for everyone!
101
Hooray!
102
We've made it to the end of another exciting stat quest!
103
If you like this StatQuest and want to see more, please subscribe.
104
And if you want to support StatQuest, consider contributing to my Patreon campaign,
105
becoming a channel member, buying one or two of my original songs or a t -shirt or a hoodie, or just donate.
106
The links are in the description below.
107
Alright, until next time, quest on!
同頻道
C11
The GELU, SiLU and SwiGLU activation functions, clearly explained!!!
Daily ConversationUniversity Students
C11
Naive Bayes, Clearly Explained!!!
Daily ConversationUniversity Students
C11
The Essential Main Ideas of Neural Networks
Business EnglishUniversity StudentsWorking Adults
C11
A Gentle Introduction to Machine Learning
Business EnglishUniversity StudentsWorking Adults
FOR YOU
推薦影片
C11
125 句
Introduction to statistical prediction models
Daily ConversationUniversity Students
C11
654 句
Stuart & Cliffe Knechtle | Do Agnostics Just Want To Have Control? | Give Me an Answer
Daily ConversationUniversity StudentsWorking Adults
C11
207 句
Go’s Big Mistake
Daily ConversationUniversity StudentsWorking Adults
C13
119 句
Mathematics and Physics | Lesson 2 | Understanding the Connection
Daily ConversationUniversity Students
C11
498 句
Give me 31 minutes, I will make PyTorch click forever
Daily ConversationUniversity Students
C11
712 句
Master AI in 60 Minutes with Vector Databases!
Daily ConversationUniversity Students
C12
78 句
Vanishing & Exploding Gradient explained | A problem resulting from backpropagation
Daily ConversationUniversity Students
C12
582 句
You've (Likely) Been Playing The Game Of Life Wrong
Daily ConversationUniversity StudentsWorking Adults
C16
51 句
1 6 What Illusions Teach Us
Daily ConversationUniversity Students
C11
244 句
1. Introduction to statistics - Lecture 1
Daily ConversationUniversity Students
本課的詞彙與口語要點
這節 C1 級別的口語課以影片“Gaussian Naive Bayes, Clearly Explained!!!”為素材。 影片中反覆出現的詞有：Troll, log, likelihood, candy, boop。 這段影片共有 107 個句子、1118 個單詞可供跟讀。 講話部分時長為 9:25。 說話人語速平穩，每分鐘約 119 個詞，很適合跟讀。 只有 77% 的單詞屬於英語最常用的 3,000 詞，詞彙難度較高。
影片中的重點詞彙
影片中最難的 15 個單詞，附發音和釋義：
| 單詞 | 發音 | 釋義 |
|---|---|---|
| likelihood 名詞 | /ˈlaɪklihʊd/ | 可能性 |
| popcorn 名詞 | /ˈpɑp.kɔɹn/ | 玉米花, 爆米花 |
| soda 名詞 | /ˈsoʊdə/ | 蘇打 /蘇打, 碳酸鈉 /碳酸鈉 |
| beep 名詞 | /biːp/ | 嘟嘟響 /嘟嘟響, 嘟嘟聲 /嘟嘟聲 |
| gram 名詞 | /ˈɡɹæm/ | 克, 公克 |
| deviation 名詞 | /ˌdiː.viˈeɪʃən/ | 偏離 /偏離 |
| classify 動詞 | /ˈklæs.əˌfaɪ/ | 分類 /分類, 分等 |
| correspond 動詞 | /ˌkoɹəˈspɑnd/ | 對應 /對應 |
| probability 名詞 | /ˌpɹɑ.bəˈbɪl.ə.ti/ | 可能性 |
| multiplication 名詞 | /ˌmʌltɪplɪˈkeɪʃən/ | 乘法 |
| shameless 形容詞 | /ˈʃeɪ̯mlɪs/ | 無恥 /無恥, 恬不知恥的 /恬不知恥的 |
| validation 名詞 | /ˌvæl.əˈdeɪ.ʃən/ | 肯定, 情緒價值 /情緒價值 |
| calculate 動詞 | /ˈkælkjʊleɪt/ | 計算 /計算, 算 |
| stat 副詞 | /stæt/ | 即刻 |
| axis 名詞 | /ˈæksɪs/ | 軸 /軸, 軸心 /軸心 |
影片中出現的短語動詞
| 單詞 | 釋義 |
|---|---|
| check out 動詞 | 退房 |
| show up 動詞 | 露面 |
影片中的語法
說話人最常用的結構，並附上影片中的原話：
| 結構 | 影片中的用法 |
|---|---|
| 定語從句 who / which + 從句 — 補充說明人或事物 | people who love · people who do · someone who does |
| 被動語態 be + 過去分詞 — 強調發生了什麼，而不是誰做的 | is named · is estimated · are called |
需要注意的發音
說話人用了 9 次縮略和弱讀形式，例如 don't, doesn't, I'm。請按聽到的簡短形式來說。
- “sh” 和 “zh” 音: deviation /ˌdiː.viˈeɪʃən/, multiplication /ˌmʌltɪplɪˈkeɪʃən/, shameless /ˈʃeɪ̯mlɪs/, validation /ˌvæl.əˈdeɪ.ʃən/
- 長單詞——注意重音位置: deviation /ˌdiː.viˈeɪʃən/, probability /ˌpɹɑ.bəˈbɪl.ə.ti/, multiplication /ˌmʌltɪplɪˈkeɪʃən/, validation /ˌvæl.əˈdeɪ.ʃən/, coordinate /koʊˈɔɹ.dəˌneɪt/
如何用這段影片練習
- 先完整聽一遍影片，不要開口，記下不認識的單詞。
- 用正常速度逐句跟讀，每句重複到你的節奏與說話人一致為止。
- 錄下自己的聲音並與原聲對比，特別注意 likelihood, popcorn, soda 這類單詞。
什麼是跟讀法？
跟讀法 (Shadowing) 是一種有科學依據的語言學習技巧，最初開發用於專業口譯員的培訓，並由多語言者Alexander Arguelles博士普及。這個方法簡單而強大：您在聽英語母語原聲的同時立即大聲重複——就像是一個延遲1-2秒緊跟說話者的影子。與被動聽力或語法練習不同，跟讀法強迫您的大腦和口腔肌肉同時處理並模模擬實的講話模式。研究表明它能顯著提高發音準確性，語調，節奏，連讀，聽力理解和口語流利度——使其成為雅思口語備考和真實英語交流最有效的方法之一。