쉐도잉 연습: Gaussian Naive Bayes, Clearly Explained!!! - 영상으로 영어 말하기 배우기
레슨 만드는 중...
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!
📺 같은 채널
C11
The GELU, SiLU and SwiGLU activation functions, clearly explained!!!
Business EnglishUniversity StudentsWorking Adults
C11
The Essential Main Ideas of Neural Networks
Business EnglishUniversity StudentsWorking Adults
C11
A Gentle Introduction to Machine Learning
Business EnglishUniversity StudentsWorking Adults
✨ 추천 영상
C111
93 문장
Naive Bayes
Business EnglishUniversity StudentsWorking Adults
C11
156 문장
SELECT COUNT (*) can impact your Backend Application Performance, here is why
Business EnglishUniversity StudentsWorking Adults
C12
405 문장
Netflix product manager mock interview and tips (with ex-Netflix Lead PM)
Business EnglishInterview EnglishUniversity Students
C11
227 문장
How to Use Google Gemini Al (Full Tutorial)
Business EnglishUniversity StudentsWorking Adults
C11
108 문장
How I Learned Procedural Generation
Business EnglishUniversity StudentsWorking Adults
B265 문장
Sheldon works with Penny at the cheesecake factory - The Big Bang Theory
Business EnglishDaily ConversationUniversity Students
C12
41 문장
Mathematical Statistics and the Theory of Probability in Casinos
Business EnglishUniversity StudentsWorking Adults
C12
86 문장
System Design: Why is Kafka Popular?
Business EnglishUniversity StudentsWorking Adults
C13
279 문장
LET THERE BE... Voltage? | Maxwell's Equation #2 Explained for Beginners
Business EnglishUniversity StudentsWorking Adults
C1480 문장
Everything Valuable from an MBA in 25 minutes
Business EnglishUniversity StudentsWorking Adults
이 레슨의 어휘와 말하기 포인트
이 C1 수준 말하기 레슨은 영상 “Gaussian Naive Bayes, Clearly Explained!!!”을(를) 바탕으로 합니다. 가장 자주 반복되는 단어는 다음과 같습니다: Troll, log, likelihood, candy, boop. 이 영상에는 섀도잉할 문장 107개와 단어 1118개가 있습니다. 말하는 구간의 길이는 9:25입니다. 화자는 분당 약 119단어의 일정한 속도로 말해서 섀도잉하기에 편한 속도입니다. 영어에서 가장 많이 쓰이는 3,000단어에 속하는 단어가 77%뿐이라 어휘가 어려운 편입니다.
이 영상의 핵심 어휘
영상에서 가장 어려운 단어 15개를 발음, 뜻과 함께 정리했습니다.
| 단어 | 발음 | 뜻 |
|---|---|---|
| likelihood 명사 | /ˈlaɪklihʊd/ | 가능성 |
| popcorn 명사 | /ˈpɑp.kɔɹn/ | 팝콘 |
| soda 명사 | /ˈsoʊdə/ | 소다, 탄산 나트륨 |
| beep 명사 | /biːp/ | 뛰뛰, 빵빵 |
| gram 명사 | /ˈɡɹæm/ | 그램 |
| correspond 동사 | /ˌkoɹəˈspɑnd/ | 일치하다, 부합하다 |
| probability 명사 | /ˌpɹɑ.bəˈbɪl.ə.ti/ | 확률 |
| multiplication 명사 | /ˌmʌltɪplɪˈkeɪʃən/ | 곱셈, 승법 |
| shameless 형용사 | /ˈʃeɪ̯mlɪs/ | 뻔뻔하다, 뻔뻔스럽다 |
| coordinate 동사 | /koʊˈɔɹ.dəˌneɪt/ | 조화시키다 |
| calculate 동사 | /ˈkælkjʊleɪt/ | 정산하다, 계산하다 |
| axis 명사 | /ˈæksɪs/ | 축 |
| multiply 동사 | /ˈmʌltɪplaɪ/ | 곱하다 |
| terminology 명사 | /ˌtɝ.məˈnɑ.lə.d͡ʒi/ | 술어학 |
| subscribe 동사 | /səbˈskɹaɪb/ | 구독하다 |
이 영상의 문법
화자가 가장 많이 쓰는 문형을 영상 속 실제 표현과 함께 정리했습니다.
| 문형 | 영상 속 표현 |
|---|---|
| 관계절 who / which + 절 — 사람이나 사물에 대한 추가 정보 | people who love · people who do · someone who does |
| 수동태 be + 과거분사 — 누가 하는지보다 무슨 일이 일어나는지에 초점 | is named · is estimated · are called |
주의할 발음
화자는 don't, doesn't, I'm 같은 축약형과 약화된 형태를 9번 사용합니다. 들리는 대로 짧게 발음하세요.
- “sh”와 “zh” 소리: deviation /ˌdiː.viˈeɪʃən/, multiplication /ˌmʌltɪplɪˈkeɪʃən/, shameless /ˈʃeɪ̯mlɪs/, validation /ˌvæl.əˈdeɪ.ʃən/
- 긴 단어 — 강세 위치에 주의: deviation /ˌdiː.viˈeɪʃən/, probability /ˌpɹɑ.bəˈbɪl.ə.ti/, multiplication /ˌmʌltɪplɪˈkeɪʃən/, validation /ˌvæl.əˈdeɪ.ʃən/, coordinate /koʊˈɔɹ.dəˌneɪt/
이 영상으로 연습하는 방법
- 먼저 말하지 않고 영상을 끝까지 듣고 모르는 단어를 적어 둡니다.
- 보통 속도로 한 문장씩 섀도잉하고, 화자의 리듬과 맞을 때까지 반복합니다.
- 자신의 목소리를 녹음해 원본과 비교하고, likelihood, popcorn, soda 같은 단어에 특히 주의합니다.
쉐도잉이란? 영어 실력을 빠르게 키우는 과학적 방법
쉐도잉(Shadowing)은 원래 전문 통역사 훈련을 위해 개발된 언어 학습 기법으로, 다언어 학자인 Dr. Alexander Arguelles에 의해 대중화된 방법입니다. 핵심 원리는 간단하지만 매우 강력합니다: 원어민의 영어를 들으면서 1~2초의 짧은 지연으로 즉시 소리 내어 따라 말하는 것——마치 '그림자(shadow)'처럼 화자를 따라가는 것입니다. 문법 공부나 수동적인 청취와 달리, 쉐도잉은 뇌와 입 근육이 동시에 실시간으로 영어를 처리하고 재현하도록 훈련합니다. 연구에 따르면 이 방법은 발음 정확도, 억양, 리듬, 연음, 청취력, 말하기 유창성을 크게 향상시킵니다. IELTS 스피킹 준비와 자연스러운 영어 소통을 원하는 분들에게 특히 효과적입니다.