En
stringlengths
1
23.4k
Ar
stringlengths
1
17.3k
EN_word_count
int64
1
4.01k
AR_word_count
int64
0
3.19k
and it has more capabilities compared  to perceptron i will talk about  
ولديها المزيد مقارنة بقدرات Perceptron ، سأتحدث عن
12
8
the rest of these subjects in the next  videos and classes thank you very much
بقية هذه الموضوعات في مقاطع الفيديو والفصول الدراسية التالية ، شكرًا جزيلاً لك
15
13
okay in the previous video we talked about  decision 3 and we said that decision 3 is a  
بخير في الفيديو السابق تحدثنا عن القرار 3 وقلنا أن القرار 3 هو مصنف
18
14
very important classifier that we  could use for classifying nominal data
مهم جدًا يمكننا استخدامه لتصنيف البيانات الاسمية
11
7
okay uh we said that in the  decision tree the learning model  
حسنًا أه قلنا أنه في شجرة القرار ، نموذج التعلم
12
10
is a tree okay the learning model is a tree  the learning model as we said before is  
عبارة عن شجرة حسنًا ، نموذج التعلم عبارة عن شجرة ، نموذج التعلم كما قلنا من قبل هو
18
18
so here we have a tree that wants to represent  the relation between the inputs and output
هنا لدينا شجرة تريد تمثيل العلاقة بين المدخلات والمخرجات حسنًا
17
10
okay the most important thing  about the decision tree is the
، أهم شيء في شجرة القرار هو
11
7
building of the tree okay building  the tree is the most important thing  
بناء الشجرة ، حسنًا ، بناء الشجرة هو أهم شيء
13
10
that we are doing for the decision tree
نقوم به لاتخاذ القرار حسنًا ، عندما نريد بناء الشجرة ، فإن أهم شيء هو
8
15
selecting the best features  to put into the tree okay  
أفضل الميزات لوضعها في الشجرة ،
10
6
so when we want to build the tree the most  important thing is selecting the best features  
لذلك عندما نريد بناء الشجرة ، فإن أهم شيء هو اختيار أفضل ميزة في
17
14
in the tree yeah okay how do we do this how do  we do this how do we select the best features in  
الشجرة ، حسنًا ، كيف نفعل هذا كيف نفعل هذا كيف نختار أفضل الميزات في
23
15
the tree okay as you saw in the previous slides uh  in this tree okay here we have some features okay  
الشجرة حسنًا كما رأيتم في الشرائح السابقة أه في هذه الشجرة حسنًا هنا لدينا بعض الميزات حسنًا ،
21
18
okay now how do we select the best trees how do we  select the best features to put in the tree okay  
حسنًا الآن كيف يمكننا حدد أفضل الأشجار ، كيف نختار أفضل الميزات لوضعها في الشجرة ، حسنًا
22
17
uh in the id3 decision tree okay we are  talking about the id3 id3 is uh is one of the  
أه في شجرة قرارات id3 حسنًا ، نحن نتحدث عن id3 id3 ، أه واحد من
20
16
decision trees that we have in id3  
أشجار القرار التي لدينا في id3
7
6
the features the best features are selected  using a metric called information gain
، أفضل الميزات هي تم التحديد باستخدام مقياس يسمى اكتساب المعلومات
13
11
okay so information gain is the metric  used for selecting the best features
جيدًا ، لذا فإن اكتساب المعلومات هو المقياس المستخدم لاختيار أفضل الميزات ،
13
13
okay now we want to know what is the  information gain what does information gain mean  
حسنًا الآن نريد أن نعرف ما هي المعلومات التي تكتسب ما الذي تكتسبه المعلومات يعني أن اكتساب
16
17
information gain is a metric which is based  on on another concept and named entropy  
المعلومات هو مقياس يعتمد على مفهوم آخر ويسمى الإنتروبيا
15
9
okay now what is entropy if we have a data  
حسنًا الآن ما هو الانتروبيا إذا كانت لدينا بيانات
10
9
s is our data the entropy of this data  could be calculated using this formula
s هي بياناتنا ، يمكن حساب إنتروبيا هذه البيانات باستخدام هذه الصيغة
15
12
okay in this formula we are doing a summation  i is equal 1 to c what is c c is the number of  
، حسنًا في هذه الصيغة ، نحن نقوم بالتجميع i يساوي 1 إلى c ما هو cc هو عدد
23
19
classes for example if we are working for  fruit classification and we have five fruits  
فئات مثال إذا كنا نعمل بالنسبة لتصنيف الفاكهة ولدينا خمس ثمار ،
15
12
then c is equal to five so for each class we  are computing this okay we are calculating  
فإن c تساوي خمسة ، لذلك لكل فئة نقوم بحسابها ، حسنًا ، نقوم بحساب
18
15
this term what is p i p i means the probability  of data being in that class p i i is the index  
هذا المصطلح ، ما هو pipi يعني احتمال وجود البيانات في تلك الفئة pii هو مؤشر
23
16
of the class a logarithm p pi again negative pi  multiplied by logarithms in the base two of p i  
الفئة a اللوغاريتم p pi مرة أخرى سالب pi مضروبًا في اللوغاريتمات في الأساس اثنين من pi
20
17
for example if we have if our data has two  classes the entropy of s is equal to this  
على سبيل المثال إذا كان لدينا فئتان ، فإن إنتروبيا s تساوي هذا
19
13
and the two classes are positive and negative  okay minus the probability of positive data  
والفئتين موجبة وسالبة على ما يرام مطروحًا منها احتمال البيانات الموجبة
15
11
or the logarithms of this and probability  of negative data i'll look at it okay
أو اللوغاريتمات من هذا واحتمالية البيانات السلبية ، سأنظر إليها ، حسنًا ،
15
13
this is entropy so if we have a data  that this data it has multiple classes  
هذا إنتروبيا ، لذا إذا كان لدينا بيانات تفيد بأن هذه البيانات تحتوي على فئات متعددة ،
16
17
we could calculate the entropy using this  formula example example if we have this data okay  
فيمكننا حساب الكون باستخدام مثال مثال الصيغة هذا إذا كانت لدينا هذه البيانات ، حسنًا
16
15
this data here you see that we this data  has eight rows one two to eight rows  
هذه البيانات هنا أنت نرى أن لدينا هذه البيانات ثمانية صفوف من صف واحد إلى ثمانية صفوف
17
17
and in this data it's about tennis the sport  tennis and this this column this column it is the  
وفي هذه البيانات تتعلق بالتنس ورياضة التنس وهذا العمود هذا العمود هو
19
12
class it is the output or the class
الفصل الذي يمثل الناتج أو الفصل
8
6
okay so in this data we  have two classes no and yes
جيدًا ، لذا في هذه البيانات لدينا فئتان لا ا d نعم
12
12
okay we have two classes no yes
حسنًا ، لدينا فئتان لا نعم
7
6
and
،
1
1
so now we want to calculate entropy we want  to calculate entropy entropy d1 to d8 d1 to d8
والآن نريد حساب الانتروبيا ، نريد حساب الانتروبيا d1 إلى d8 d1 إلى d8
19
14
okay we want to calculate the entropy  
حسنًا ، نريد حساب الانتروبيا
7
5
uh we say that okay positive it means  yes positive is yes and negative is no
آه نقول إنه حسنًا إيجابي فهذا يعني نعم الموجب هو نعم وسالب لا
16
13
okay this negative is no and positive is  yes minus the probability of positive okay  
بأس بهذا السالب لا والإيجابي نعم ناقص احتمال الموجب حسنًا
15
10
positive here the probability is the number  of positive to the whole okay we have six  
موجب هنا ، الاحتمال هو عدد الموجب إلى الكل حسنًا ، لدينا ستة
16
13
positives and the whole of an eight so it's six  divided by 8 this is the probability of positive  
موجبات وكل ثمانية ، لذا فهي ستة مقسومة على 8 وهذا هو الاحتمال من موجب
19
15
and then logarithms this probability and  then minus the probability of negative  
ثم لوغاريتمات هذا الاحتمال ثم ناقص احتمالية وجود سالب
12
9
a logarithm the probability of negative okay  
لوغاريتم واحتمال الموافقة السالبة
7
4
and this will come to this value point 24  this is the entropy of this data point 24.
وهذا سيصل إلى نقطة القيمة 24 هذه هي الانتروبيا لنقطة البيانات هذه 24.
18
13
okay the best value for entropy is zero  if the entropy is zero it means that  
حسنًا ، أفضل قيمة للإنتروبيا هي صفر إذا كان الإنتروبيا تساوي صفرًا ، فهذا يعني أن
16
16
all of the data have the same class okay if all  of the data for example are yes the output of  
جميع البيانات لها نفس الفئة ، حسنًا إذا كانت جميع البيانات على سبيل المثال هي نعم ، فإن إخراج
21
19
all of the data is yes what happens here you see  that this would be uh minus eight divided by eight  
جميع البيانات هو نعم ، ما يحدث هنا ترى أن هذا سيكون أه ناقص ثمانية مقسومًا على ثمانية
21
18
okay look at eight divided by eight okay logarithm  eight divided by eight it means logarithm one and  
حسنًا انظر إلى ثمانية مقسومًا على ثمانية لوغاريتم ثمانية مقسومًا على ثمانية ، فهذا يعني أن لوغاريتم واحد
18
18
logarithm one is equal to zero and here the  probability of the negative class it is zero  
ولوغاريتم واحد يساوي صفرًا ، وهنا يكون احتمال الفئة السالبة صفرًا
17
11
so the entropy would be zero okay  this is the best situation of entropy  
، لذا فإن الانتروبيا ستكون صفرًا ، وهذا أفضل وضع للإنتروبيا
14
11
when it is zero and the bet in and the worst  situation of entropy is a one so entropy is a  
عندما تساوي صفرًا ، والرهان في وأسوأ حالة للإنتروبيا هو واحد ، لذا فإن الإنتروبيا هي
21
16
value between zero and one the worst value is one  when does this occur it occurs when we have the  
قيمة بين صفر وواحد أسوأ قيمة عندما يحدث هذا يحدث عندما يكون لدينا
20
13
same number of data in different classes for  example here we have eight data four of them  
نفس العدد من البيانات في فئات مختلفة على سبيل المثال نحن هنا لديك ثمانية بيانات ، أربعة منها
17
18
is no and four of them is yes if you calculate  this you will see that the entropy would be one
لا ، وأربعة منها نعم إذا قمت بحساب هذا ، سترى أن الانتروبيا ستكون واحدة على
21
16
okay okay so this is entropy
ما يرام ، لذلك هذا إنتروبيا ،
6
7
this is entropy a value that we have
هذا هو إنتروبيا لدينا
8
4
here
هنا
1
1
now
الآن
1
1
we said that in id3 information gain  is used to select the best features
قلنا أنه في اكتساب المعلومات id3 تُستخدم لتحديد أفضل الميزات ،
14
11
okay now what is information gain  this is the formula of information  
حسنًا الآن ما هو اكتساب المعلومات هذه هي معادلة
12
9
gain sorry this is the formula of information  gain if we have a data s and we have an attribute  
اكتساب المعلومات آسف هذه هي صيغة كسب المعلومات إذا كان لدينا بيانات ولدينا سمة
20
14
a the information gain of attribute a in  data s is equal to entropy s minus this value
اكتساب المعلومات من السمة في البيانات هي إكوا l إلى entropy s ناقص هذه القيمة
18
15
okay
، حسنًا
1
2
so the information gain of an  attribute could be calculated  
، يمكن حساب كسب المعلومات الخاص بسمة ما
10
8
using this term okay first we have to  calculate the entropy and then we should  
باستخدام هذا المصطلح ، حسنًا أولاً علينا حساب الانتروبيا ثم علينا
15
11
calculate the entropy for the attribute  okay let's look and add an example of this  
حساب الانتروبيا للسمة ، حسنًا ، دعنا ننظر ونضيف مثالًا على هذا
15
12
how could we calculate the information gain for  a feature for okay again assume we have this data  
كيف يمكن نحسب كسب المعلومات لميزة حسناً مرة أخرى ، نفترض أن لدينا هذه البيانات
18
15
okay we have this data and in this data what  do we want to do we want to calculate the gain  
، حسنًا ، لدينا هذه البيانات وفي هذه البيانات ، ماذا نريد أن نفعل نحن نريد حساب كسب
21
18
of the feature wind okay this feature we  want to calculate the gain the information  
الرياح الميزة ، حسنًا ، هذه الميزة نريد حساب الكسب اكتساب المعلومات
15
12
gain of the feature wind in this data s is this  data that we have here okay s is this state
لرياح الميزة في هذه البيانات هي هذه البيانات التي لدينا هنا حسنًا ، هل هذه الحالة
21
16
okay so we want to calculate the  information gain of the feature wind okay
جيدة لذلك نريد حساب كسب المعلومات لرياح الميزة بشكل جيد
14
10
in this slide you can see that we  are calculating the information gain
في هذه الشريحة ، يمكنك أن ترى أننا نحسب اكتساب المعلومات
13
11
okay
حسنًا
1
1
you see that for this feature the feature wind
، ترى أنه بالنسبة لهذه الميزة ، فإن ميزة الرياح
9
10
we have this value this is the values of  future wind the feature wind the values  
لدينا هذه القيمة ، فهذه هي قيم الرياح المستقبلية ، حيث تعمل الميزة على أن تكون القيم
16
17
are weak or strong as you see this feature is  a nominal feature two values are string values  
ضعيفة أو قوية كما ترى هذه الميزة هي ميزة اسمية قيمتان عبارة عن قيم سلسلة
18
15
and the number of values is limited only  two values okay now in s is is the data  
و عدد القيم محدود فقط قيمتان حسنًا الآن في s هي البيانات
18
12
in the previous slide you see that this  here we have 14 data d1 d14 14 data okay  
الموجودة في الشريحة السابقة التي تراها هنا لدينا 14 بيانات d1 d14 14 بيانات جيدة
18
15
now in this 14 data nine of them is positive  and five is negative if you look and a yes means  
الآن في هذه 14 بيانات تسعة منها موجبة وخمسة سالبة إذا نظرتم ونعم تعني
21
14
positive and no means negative and 9 is positive  and 5 is negative okay then if we look in the data  
موجب ولا يعني سلبي و 9 موجب و 5 سلبي حسنًا ، إذا نظرنا في البيانات
21
16
which the wind feature has the value weak in this  data we have six positive and two negative and in  
التي ميزة الرياح لها قيمة ضعيفة في هذه البيانات لدينا ستة موجبة واثنتان سلبية وفي
20
15
the data which the wind feature has the value  is strong we have three positive three negative
البيانات التي ميزة الرياح لها قيمة قوية ولدينا ثلاثة موجب ثلاثة سلبي
17
12
okay and this is the gain the gain of the feature  wind in data s it's equal to entropy s entropy s  
حسنًا وهذا هو كسب ميزة الرياح في البيانات s التي تساوي الانتروبيا s
22
13
it could be calculated using this data okay and  the formula that we have in the previous slides  
التي يمكن حسابها باستخدام هذه البيانات حسنًا والمعادلة التي لدينا في الشرائح السابقة
18
13
and then uh this this term okay entropy is the  value of entropy s is point 94 and this one  
ثم أه هذا المصطلح حسنًا ، الإنتروبيا هي قيمة الإنتروبيا s هي النقطة 94 وهذا
20
15
is weak s weak means the data which the wind  feature is weak we said six positive and two  
ضعيف s ضعيف يعني البيانات التي تكون ميزة الرياح ضعيفة قلناها ستة موجبة واثنين
19
14
negative okay so this is the entropy we put  it here and the value of the entropy is 0.8  
سلبيين حسنًا ، هذا هو الإنتروبيا نضعها هنا وقيمة الإدخال opy هو 0.8
19
13
and for s is strong it means the data that  the wind feature it has the value is strong  
ولأن s قوي فهذا يعني أن البيانات التي تحتوي على ميزة الرياح هي قوية
19
14
the entropy is one okay okay now what is  this eight divided by 14 it means that it  
وأن الانتروبيا هي واحدة على ما يرام الآن ما هذه الثمانية مقسومة على 14 فهذا يعني
18
16
means that from the all of the data that we have  we have 14 data from this 14 data eight of them  
أنه من جميع البيانات هل لدينا 14 بيانات من هذه البيانات الأربعة عشر ، ثمانية منها
22
16
have the wind value had the wind feature has the  value weak and six of them the wind feature has  
لها قيمة الرياح ، وكانت قيمة الرياح ضعيفة وستة منها ميزة الرياح لها
20
13
the value is strong so we have written this this  8 divided by 14 it is this term in the formula  
قيمة قوية ، لذلك كتبنا هذا 8 مقسومًا على 14 هذا هو هذا المصطلح في الصيغة ،
21
17
okay so the gain of the feature  wind finally is this point .04
حسنًا ، فإن كسب ميزة الرياح أخيرًا هو هذه النقطة .04
13
11
okay this is the gain for  the feature wind in the data
حسنًا ، هذا هو مكسب الرياح المميزة في البيانات
12
9
so we could calculate the gain  for different features and then  
حتى نتمكن من حساب الكسب لميزات مختلفة ومن ثم
11
9
we could select the best feature the best  feature is the feature which has the highest gain  
يمكننا تحديد أفضل ميزة أفضل ميزة هي الميزة التي لها أعلى مكاسب
17
12