4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

DonorsChoose
DonorsChoose.org receives hundreds of thousands of project proposals each year for classroom projects in need of funding. Right now, a large
number of volunteers is needed to manually screen each submission before it's approved to be posted on the DonorsChoose.org website.

Next year, DonorsChoose.org expects to receive close to 500,000 project proposals. As a result, there are three main problems they need to solve:

How to scale current manual processes and resources to screen 500,000 projects so that they can be posted as quickly and as efficiently as
possible
How to increase the consistency of project vetting across different volunteers to improve the experience for teachers
How to focus volunteer time on the applications that need the most assistance

The goal of the competition is to predict whether or not a DonorsChoose.org project proposal submitted by a teacher will be approved, using the text
of project descriptions as well as additional metadata about the project, teacher, and school. DonorsChoose.org can then use this information to
identify projects most likely to need further review before approval.

1 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

About the DonorsChoose Data Set

The train.csv data set provided by DonorsChoose contains the following features:

Feature Description

project_id A unique identifier for the proposed project. Example: p036502

Title of the project. Examples:

project_title Art Will Make You Happy!

First Grade Fun

Grade level of students for which the project is targeted. One of the following enumerated values:

Grades PreK-2
project_grade_category Grades 3-5
Grades 6-8
Grades 9-12

One or more (comma-separated) subject categories for the project from the following enumerated list of
values:

Applied Learning
Care & Hunger
Health & Sports
History & Civics
Literacy & Language
Math & Science
project_subject_categories Music & The Arts
Special Needs
Warmth

Examples:

Music & The Arts

Literacy & Language, Math & Science

State where school is located (Two-letter U.S. postal code (https://en.wikipedia.org

school_state
/wiki/List_of_U.S._state_abbreviations#Postal_codes)). Example: WY

One or more (comma-separated) subject subcategories for the project. Examples:

project_subject_subcategories Literacy
Literature & Writing, Social Sciences

An explanation of the resources needed for the project. Example:

project_resource_summary My students need hands on literacy materials to manage sensory needs!

</code

project_essay_1 First application essay*

project_essay_2 Second application essay*

project_essay_3 Third application essay*

project_essay_4 Fourth application essay*

project_submitted_datetime Datetime when project application was submitted. Example: 2016-04-28 12:43:56.245

A unique identifier for the teacher of the proposed project. Example:

teacher_id
bdf8baa8fedef6bfeec7ae4ff1c15c56

Teacher's title. One of the following enumerated values:

nan
Dr.
teacher_prefix Mr.
Mrs.
Ms.
Teacher.

teacher_number_of_previously_posted_projects Number of project applications previously submitted by the same teacher. Example: 2

* See the section Notes on the Essay Data for more details about these features.

Additionally, the resources.csv data set provides more data about the resources required for each project. Each line in this file represents a
resource required by a project:

Feature Description

id A project_id value from the train.csv file. Example: p036502

description Desciption of the resource. Example: Tenor Saxophone Reeds, Box of 25

quantity Quantity of the resource required. Example: 3

price Price of the resource required. Example: 9.95

Note: Many projects require multiple resources. The id value corresponds to a project_id in train.csv, so you use it as a key to retrieve all

2 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

Notes on the Essay Data

Prior to May 17, 2016, the prompts for the essays were as follows:
__project_essay_1:__ "Introduce us to your classroom"
__project_essay_2:__ "Tell us more about your students"
__project_essay_3:__ "Describe how your students will use the materials you're requesting"
__project_essay_3:__ "Close by sharing why your project will make a difference"

Starting on May 17, 2016, the number of essays was reduced from 4 to 2, and the prompts for the first 2 essays were changed to the following:
__project_essay_1:__ "Describe your students: What makes your students special? Specific details about their background, your
neighborhood, and your school are all helpful."
__project_essay_2:__ "About your project: How will these materials make a difference in your students' learning and improve their school
lives?"

For all projects with project_submitted_datetime of 2016-05-17 and later, the values of project_essay_3 and project_essay_4 will be NaN.

In [208]: %matplotlib inline

import warnings
warnings.filterwarnings("ignore")

import sqlite3
import pandas as pd
import numpy as np
import nltk
import string
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import TfidfVectorizer

from sklearn.feature_extraction.text import CountVectorizer

from sklearn.metrics import confusion_matrix
from sklearn import metrics
from sklearn.metrics import roc_curve, auc
from nltk.stem.porter import PorterStemmer

import re
# Tutorial about Python regular expressions: https://pymotw.com/2/re/
import string
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer
from nltk.stem.wordnet import WordNetLemmatizer

from gensim.models import Word2Vec

from gensim.models import KeyedVectors
import pickle

from tqdm import tqdm

import os

from plotly import plotly

import plotly.offline as offline
import plotly.graph_objs as go
offline.init_notebook_mode()
from collections import Counter
import gc

1.1 Reading Data

In [209]: project_data = pd.read_csv('train_data.csv')
resource_data = pd.read_csv('resources.csv')
In [210]: print("Number of data points in train data", project_data.shape)
print('-'*50)
print("The attributes of data :", project_data.columns.values)
Number of data points in train data (109248, 17)
--------------------------------------------------
The attributes of data : ['Unnamed: 0' 'id' 'teacher_id' 'teacher_prefix' 'school_state'
'project_submitted_datetime' 'project_grade_category'
'project_subject_categories' 'project_subject_subcategories'
'project_title' 'project_essay_1' 'project_essay_2' 'project_essay_3'
'project_essay_4' 'project_resource_summary'
'teacher_number_of_previously_posted_projects' 'project_is_approved']

3 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [211]: print("Number of data points in train data", resource_data.shape)

print(resource_data.columns.values)
resource_data.head(2)
Number of data points in train data (1541272, 4)
['id' 'description' 'quantity' 'price']

Out[211]:
id description quantity price

0 p233245 LC652 - Lakeshore Double-Space Mobile Drying Rack 1 149.00

1 p069063 Bouncy Bands for Desks (Blue support pipes) 3 14.95

1.2 preprocessing of project_subject_categories

In [212]: catogories = list(project_data['project_subject_categories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039

# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
cat_list = []
for i in catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&",
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The'
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science
temp+=j.strip()+" " #" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_') # we are replacing the & value into
cat_list.append(temp.strip())

project_data['clean_categories'] = cat_list
project_data.drop(['project_subject_categories'], axis=1, inplace=True)

from collections import Counter

my_counter = Counter()
for word in project_data['clean_categories'].values:
my_counter.update(word.split())

cat_dict = dict(my_counter)
sorted_cat_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))

1.3 preprocessing of project_subject_subcategories

In [213]: sub_catogories = list(project_data['project_subject_subcategories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039

# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python

sub_cat_list = []
for i in sub_catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&",
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The'
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science
temp +=j.strip()+" "#" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_')
sub_cat_list.append(temp.strip())

project_data['clean_subcategories'] = sub_cat_list
project_data.drop(['project_subject_subcategories'], axis=1, inplace=True)

# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039

my_counter = Counter()
for word in project_data['clean_subcategories'].values:
my_counter.update(word.split())

sub_cat_dict = dict(my_counter)
sorted_sub_cat_dict = dict(sorted(sub_cat_dict.items(), key=lambda kv: kv[1]))

4 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

1.3 Text preprocessing

In [214]: # merge two column text dataframe:
project_data["essay"] = project_data["project_essay_1"].map(str) +\
project_data["project_essay_2"].map(str) + \
project_data["project_essay_3"].map(str) + \
project_data["project_essay_4"].map(str)
In [215]: project_data.head(2)

Out[215]:
Unnamed:
id teacher_id teacher_prefix school_state project_submitted_datetime project_grade_category project_title
0

Educational
Support for
0 160221 p253737 c90749f5d961ff158d4b4d1e7dc665fc Mrs. IN 2016-12-05 13:43:57 Grades PreK-2
Learners at

Projector for
1 140945 p258326 897464ce9ddc600bced1151f324dd63a Mr. FL 2016-10-25 09:22:10 Grades 6-8
Learners

In [216]: #### 1.4.2.3 Using Pretrained Models: TFIDF weighted W2V

5 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [217]: # printing some random reviews

print(project_data['essay'].values[0])
print("="*50)
print(project_data['essay'].values[150])
print("="*50)
print(project_data['essay'].values[1000])
print("="*50)
print(project_data['essay'].values[20000])
print("="*50)
print(project_data['essay'].values[99999])
print("="*50)
My students are English learners that are working on English as their second or third languages. We are a
melting pot of refugees, immigrants, and native-born Americans bringing the gift of language to our schoo
l. \r\n\r\n We have over 24 languages represented in our English Learner program with students at every l
evel of mastery. We also have over 40 countries represented with the families within our school. Each s
tudent brings a wealth of knowledge and experiences to us that open our eyes to new cultures, beliefs, an
d respect.\"The limits of your language are the limits of your world.\"-Ludwig Wittgenstein Our English
learner's have a strong support system at home that begs for more resources. Many times our parents are
learning to read and speak English along side of their children. Sometimes this creates barriers for par
ents to be able to help their child learn phonetics, letter recognition, and other reading skills.\r\n\r\
nBy providing these dvd's and players, students are able to continue their mastery of the English languag
e even if no one at home is able to assist. All families with students within the Level 1 proficiency st
atus, will be a offered to be a part of this program. These educational videos will be specially chosen
by the English Learner Teacher and will be sent home regularly to watch. The videos are to help the chil
d develop early reading skills.\r\n\r\nParents that do not have access to a dvd player will have the oppo
rtunity to check out a dvd player to use for the year. The plan is to use these videos and educational d
vd's for the years to come for other EL students.\r\nnannan
==================================================
The 51 fifth grade students that will cycle through my classroom this year all love learning, at least mo
st of the time. At our school, 97.3% of the students receive free or reduced price lunch. Of the 560 stud
ents, 97.3% are minority students. \r\nThe school has a vibrant community that loves to get together and
celebrate. Around Halloween there is a whole school parade to show off the beautiful costumes that studen
ts wear. On Cinco de Mayo we put on a big festival with crafts made by the students, dances, and games. A
t the end of the year the school hosts a carnival to celebrate the hard work put in during the school yea
r, with a dunk tank being the most popular activity.My students will use these five brightly colored Hokk
i stools in place of regular, stationary, 4-legged chairs. As I will only have a total of ten in the clas
sroom and not enough for each student to have an individual one, they will be used in a variety of ways.
During independent reading time they will be used as special chairs students will each use on occasion. I
will utilize them in place of chairs at my small group tables during math and reading times. The rest of
the day they will be used by the students who need the highest amount of movement in their life in order
to stay focused on school.\r\n\r\nWhenever asked what the classroom is missing, my students always say mo
re Hokki Stools. They can't get their fill of the 5 stools we already have. When the students are sitting
in group with me on the Hokki Stools, they are always moving, but at the same time doing their work. Anyt
ime the students get to pick where they can sit, the Hokki Stools are the first to be taken. There are al
ways students who head over to the kidney table to get one of the stools who are disappointed as there ar
e not enough of them. \r\n\r\nWe ask a lot of students to sit for 7 hours a day. The Hokki stools will be
a compromise that allow my students to do desk work and move at the same time. These stools will help stu
dents to meet their 60 minutes a day of movement by allowing them to activate their core muscles for bala
nce while they sit. For many of my students, these chairs will take away the barrier that exists in schoo
ls for a child who can't sit still.nannan
==================================================
How do you remember your days of school? Was it in a sterile environment with plain walls, rows of desks,
and a teacher in front of the room? A typical day in our room is nothing like that. I work hard to create
a warm inviting themed room for my students look forward to coming to each day.\r\n\r\nMy class is made u
p of 28 wonderfully unique boys and girls of mixed races in Arkansas.\r\nThey attend a Title I school, wh
ich means there is a high enough percentage of free and reduced-price lunch to qualify. Our school is an
\"open classroom\" concept, which is very unique as there are no walls separating the classrooms. These 9
and 10 year-old students are very eager learners; they are like sponges, absorbing all the information an
d experiences and keep on wanting more.With these resources such as the comfy red throw pillows and the w
himsical nautical hanging decor and the blue fish nets, I will be able to help create the mood in our cla
ssroom setting to be one of a themed nautical environment. Creating a classroom environment is very impor
tant in the success in each and every child's education. The nautical photo props will be used with each
child as they step foot into our classroom for the first time on Meet the Teacher evening. I'll take pict
ures of each child with them, have them developed, and then hung in our classroom ready for their first d
ay of 4th grade. This kind gesture will set the tone before even the first day of school! The nautical t
hank you cards will be used throughout the year by the students as they create thank you cards to their t
eam groups.\r\n\r\nYour generous donations will help me to help make our classroom a fun, inviting, learn
ing environment from day one.\r\n\r\nIt costs lost of money out of my own pocket on resources to get our
classroom ready. Please consider helping with this project to make our new school year a very successful
one. Thank you!nannan
==================================================
My kindergarten students have varied disabilities ranging from speech and language delays, cognitive dela
ys, gross/fine motor delays, to autism. They are eager beavers and always strive to work their hardest wo
rking past their limitations. \r\n\r\nThe materials we have are the ones I seek out for my students. I te
ach in a Title I school where most of the students receive free or reduced price lunch. Despite their di
sabilities and limitations, my students love coming to school and come eager to learn and explore.Have yo
u ever felt like you had ants in your pants and you needed to groove and move as you were in a meeting? T
his is how my kids feel all the time. The want to be able to move as they learn or so they say.Wobble cha
irs are the answer and I love then because they develop their core, which enhances gross motor and in Tur
n fine motor skills. \r\nThey also want to learn through games, my kids don't want to sit and do workshee
ts. They want to learn to count by jumping and playing. Physical engagement is the key to our success. Th

6 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [218]: # https://stackoverflow.com/a/47091490/4084039
import re

def decontracted(phrase):
# specific
phrase = re.sub(r"won't", "will not", phrase)
phrase = re.sub(r"can\'t", "can not", phrase)

# general
phrase = re.sub(r"n\'t", " not", phrase)
phrase = re.sub(r"\'re", " are", phrase)
phrase = re.sub(r"\'s", " is", phrase)
phrase = re.sub(r"\'d", " would", phrase)
phrase = re.sub(r"\'ll", " will", phrase)
phrase = re.sub(r"\'t", " not", phrase)
phrase = re.sub(r"\'ve", " have", phrase)
phrase = re.sub(r"\'m", " am", phrase)
return phrase
In [219]: sent = decontracted(project_data['essay'].values[20000])
print(sent)
print("="*50)
My kindergarten students have varied disabilities ranging from speech and language delays, cognitive dela
ys, gross/fine motor delays, to autism. They are eager beavers and always strive to work their hardest wo
rking past their limitations. \r\n\r\nThe materials we have are the ones I seek out for my students. I te
ach in a Title I school where most of the students receive free or reduced price lunch. Despite their di
sabilities and limitations, my students love coming to school and come eager to learn and explore.Have yo
u ever felt like you had ants in your pants and you needed to groove and move as you were in a meeting? T
his is how my kids feel all the time. The want to be able to move as they learn or so they say.Wobble cha
irs are the answer and I love then because they develop their core, which enhances gross motor and in Tur
n fine motor skills. \r\nThey also want to learn through games, my kids do not want to sit and do workshe
ets. They want to learn to count by jumping and playing. Physical engagement is the key to our success. T
he number toss and color and shape mats can make that happen. My students will forget they are doing work
and just have the fun a 6 year old deserves.nannan
==================================================

In [220]: # \r \n \t remove from string python: http://texthandler.com/info/remove-line-breaks-python/

sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
print(sent)
My kindergarten students have varied disabilities ranging from speech and language delays, cognitive dela
ys, gross/fine motor delays, to autism. They are eager beavers and always strive to work their hardest wo
rking past their limitations. The materials we have are the ones I seek out for my students. I teach
in a Title I school where most of the students receive free or reduced price lunch. Despite their disabi
lities and limitations, my students love coming to school and come eager to learn and explore.Have you ev
er felt like you had ants in your pants and you needed to groove and move as you were in a meeting? This
is how my kids feel all the time. The want to be able to move as they learn or so they say.Wobble chairs
are the answer and I love then because they develop their core, which enhances gross motor and in Turn fi
ne motor skills. They also want to learn through games, my kids do not want to sit and do worksheets. T
hey want to learn to count by jumping and playing. Physical engagement is the key to our success. The num
ber toss and color and shape mats can make that happen. My students will forget they are doing work and j
ust have the fun a 6 year old deserves.nannan

In [221]: #remove spacial character: https://stackoverflow.com/a/5843547/4084039

sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
print(sent)
My kindergarten students have varied disabilities ranging from speech and language delays cognitive delay
s gross fine motor delays to autism They are eager beavers and always strive to work their hardest workin
g past their limitations The materials we have are the ones I seek out for my students I teach in a Title
I school where most of the students receive free or reduced price lunch Despite their disabilities and li
mitations my students love coming to school and come eager to learn and explore Have you ever felt like y
ou had ants in your pants and you needed to groove and move as you were in a meeting This is how my kids
feel all the time The want to be able to move as they learn or so they say Wobble chairs are the answer a
nd I love then because they develop their core which enhances gross motor and in Turn fine motor skills T
hey also want to learn through games my kids do not want to sit and do worksheets They want to learn to c
ount by jumping and playing Physical engagement is the key to our success The number toss and color and s
hape mats can make that happen My students will forget they are doing work and just have the fun a 6 year
old deserves nannan

7 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [222]: # https://gist.github.com/sebleier/554280
# we are removing the words from the stop words list: 'no', 'nor', 'not'
stopwords= ['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're", "you've",\
"you'll", "you'd", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \
'she', "she's", 'her', 'hers', 'herself', 'it', "it's", 'its', 'itself', 'they', 'them', 'their'
'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', "that'll", 'these', 'those'
'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do',
'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while',
'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before'
'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again'
'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few'
'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \
's', 't', 'can', 'will', 'just', 'don', "don't", 'should', "should've", 'now', 'd', 'll', 'm',
've', 'y', 'ain', 'aren', "aren't", 'couldn', "couldn't", 'didn', "didn't", 'doesn', "doesn't",
"hadn't", 'hasn', "hasn't", 'haven', "haven't", 'isn', "isn't", 'ma', 'mightn', "mightn't", 'mustn'
"mustn't", 'needn', "needn't", 'shan', "shan't", 'shouldn', "shouldn't", 'wasn', "wasn't", 'weren'
'won', "won't", 'wouldn', "wouldn't"]

In [223]: # Combining all the above stundents

from tqdm import tqdm
preprocessed_essays = []
# tqdm is for printing the status bar
for sentance in tqdm(project_data['essay'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
preprocessed_essays.append(sent.lower().strip())
100%|████████████████████████████████████████████████████████████████████████| 109248/109248 [01:11<00:00
, 1527.95it/s]

In [224]: # after preprocesing

preprocessed_essays[20000]
Out[224]: 'my kindergarten students varied disabilities ranging speech language delays cognitive delays gross fine
motor delays autism they eager beavers always strive work hardest working past limitations the materials
ones i seek students i teach title i school students receive free reduced price lunch despite disabilitie
s limitations students love coming school come eager learn explore have ever felt like ants pants needed
groove move meeting this kids feel time the want able move learn say wobble chairs answer i love develop
core enhances gross motor turn fine motor skills they also want learn games kids not want sit worksheets
they want learn count jumping playing physical engagement key success the number toss color shape mats ma
ke happen my students forget work fun 6 year old deserves nannan'

1.4 Preprocessing of `project_title`

In [225]: # similarly you can preprocess the titles also
from tqdm import tqdm
preprocessed_title = []
# tqdm is for printing the status bar
for sentance in tqdm(project_data['project_title'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
preprocessed_title.append(sent.lower().strip())
100%|███████████████████████████████████████████████████████████████████████| 109248/109248 [00:03<00:00,
33673.70it/s]

1.5 Preparing data for models

In [226]: project_data.columns

Out[226]: Index(['Unnamed: 0', 'id', 'teacher_id', 'teacher_prefix', 'school_state',

'project_submitted_datetime', 'project_grade_category', 'project_title',
'project_essay_1', 'project_essay_2', 'project_essay_3',
'project_essay_4', 'project_resource_summary',
'teacher_number_of_previously_posted_projects', 'project_is_approved',
'clean_categories', 'clean_subcategories', 'essay'],
dtype='object')

8 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

we are going to consider

- school_state : categorical data

- clean_categories : categorical data
- clean_subcategories : categorical data
- project_grade_category : categorical data
- teacher_prefix : categorical data

- project_title : text data

- text : text data
- project_resource_summary: text data (optinal)

- quantity : numerical (optinal)

- teacher_number_of_previously_posted_projects : numerical
- price : numerical

1.5.1 Vectorizing Categorical data

https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/handling-categorical-and-numerical-features/
(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/handling-categorical-and-numerical-features/)

In [227]: # # we use count vectorizer to convert the values into one

# from sklearn.feature_extraction.text import CountVectorizer
# vectorizer = CountVectorizer(vocabulary=list(sorted_cat_dict.keys()), lowercase=False, binary=True)
# categories_one_hot = vectorizer.fit_transform(project_data['clean_categories'].values)
# print(vectorizer.get_feature_names())
# print("Shape of matrix after one hot encodig ",categories_one_hot.shape)
In [228]: # we use count vectorizer to convert the values into one
# vectorizer = CountVectorizer(vocabulary=list(sorted_sub_cat_dict.keys()), lowercase=False, binary=True)
# sub_categories_one_hot = vectorizer.fit_transform(project_data['clean_subcategories'].values)
# print(vectorizer.get_feature_names())
# print("Shape of matrix after one hot encodig ",sub_categories_one_hot.shape)
In [229]: # you can do the similar thing with state, teacher_prefix and project_grade_category also

1.5.2 Vectorizing Text data

1.5.2.1 Bag of words

In [230]: # We are considering only the words which appeared in at least 10 documents(rows or projects).
# vectorizer = CountVectorizer(min_df=10)
# text_bow = vectorizer.fit_transform(preprocessed_essays)
# print("Shape of matrix after one hot encodig ",text_bow.shape)
In [231]: # you can vectorize the title also
# before you vectorize the title make sure you preprocess it

1.5.2.2 TFIDF vectorizer

In [232]: # from sklearn.feature_extraction.text import TfidfVectorizer

# vectorizer = TfidfVectorizer(min_df=10)
# text_tfidf = vectorizer.fit_transform(preprocessed_essays)
# print("Shape of matrix after one hot encodig ",text_tfidf.shape)

1.5.2.3 Using Pretrained Models: Avg W2V

9 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [233]: '''
# Reading glove vectors in python: https://stackoverflow.com/a/38230349/4084039
def loadGloveModel(gloveFile):
print ("Loading Glove Model")
f = open(gloveFile,'r', encoding="utf8")
model = {}
for line in tqdm(f):
splitLine = line.split()
word = splitLine[0]
embedding = np.array([float(val) for val in splitLine[1:]])
model[word] = embedding
print ("Done.",len(model)," words loaded!")
return model
model = loadGloveModel('glove.42B.300d.txt')

# ============================
Output:

Loading Glove Model

1917495it [06:32, 4879.69it/s]
Done. 1917495 words loaded!

# ============================

words = []
for i in preproced_texts:
words.extend(i.split(' '))

for i in preproced_titles:
words.extend(i.split(' '))
print("all the words in the coupus", len(words))
words = set(words)
print("the unique words in the coupus", len(words))

inter_words = set(model.keys()).intersection(words)
print("The number of words that are present in both glove vectors and our coupus", \
len(inter_words),"(",np.round(len(inter_words)/len(words)*100,3),"%)")

words_courpus = {}
words_glove = set(model.keys())
for i in words:
if i in words_glove:
words_courpus[i] = model[i]
print("word 2 vec length", len(words_courpus))

# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variab

import pickle
with open('glove_vectors', 'wb') as f:
pickle.dump(words_courpus, f)

'''

Out[233]: '\n# Reading glove vectors in python: https://stackoverflow.com/a/38230349/4084039\ndef (https://stackove

rflow.com/a/38230349/4084039\ndef) loadGloveModel(gloveFile):\n print ("Loading Glove Model")\n f =
open(gloveFile,\'r\', encoding="utf8")\n model = {}\n for line in tqdm(f):\n splitLine = lin
e.split()\n word = splitLine[0]\n embedding = np.array([float(val) for val in splitLine[1:]
])\n model[word] = embedding\n print ("Done.",len(model)," words loaded!")\n return model\nm
odel = loadGloveModel(\'glove.42B.300d.txt\')\n\n# ============================\nOutput:\n \nLoading G
love Model\n1917495it [06:32, 4879.69it/s]\nDone. 1917495 words loaded!\n\n# ===========================
=\n\nwords = []\nfor i in preproced_texts:\n words.extend(i.split(\' \'))\n\nfor i in preproced_titles
:\n words.extend(i.split(\' \'))\nprint("all the words in the coupus", len(words))\nwords = set(words)
\nprint("the unique words in the coupus", len(words))\n\ninter_words = set(model.keys()).intersection(wor
ds)\nprint("The number of words that are present in both glove vectors and our coupus", len(inter_w
ords),"(",np.round(len(inter_words)/len(words)*100,3),"%)")\n\nwords_courpus = {}\nwords_glove = set(mode
l.keys())\nfor i in words:\n if i in words_glove:\n words_courpus[i] = model[i]\nprint("word 2
vec length", len(words_courpus))\n\n\n# stronging variables into pickle files python: http://www.jessicay
ung.com/how-to-use-pickle-to-save-and-load-variables-in-python/\n\nimport (http://www.jessicayung.com/how
-to-use-pickle-to-save-and-load-variables-in-python/\n\nimport) pickle\nwith open(\'glove_vectors\', \'wb
\') as f:\n pickle.dump(words_courpus, f)\n\n\n'

In [234]: # stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variab

# make sure you have the glove_vectors file
# with open('glove_vectors', 'rb') as f:
# model = pickle.load(f)
# glove_words = set(model.keys())

10 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [235]: # average Word2Vec

# compute average word2vec for each review.
# avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
# for sentence in tqdm(preprocessed_essays): # for each review/sentence
# vector = np.zeros(300) # as word vectors are of zero length
# cnt_words =0; # num of words with a valid vector in the sentence/review
# for word in sentence.split(): # for each word in a review/sentence
# if word in glove_words:
# vector += model[word]
# cnt_words += 1
# if cnt_words != 0:
# vector /= cnt_words
# avg_w2v_vectors.append(vector)

# print(len(avg_w2v_vectors))
# print(len(avg_w2v_vectors[0]))

1.5.2.3 Using Pretrained Models: TFIDF weighted W2V

In [236]: # # S = ["abc def pqr", "def def def abc", "pqr pqr def"]
# tfidf_model = TfidfVectorizer()
# tfidf_model.fit(preprocessed_essays)
# # we are converting a dictionary with word as a key, and the idf as a value
# dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
# tfidf_words = set(tfidf_model.get_feature_names())
In [237]: # # average Word2Vec
# # compute average word2vec for each review.
# tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
# for sentence in tqdm(preprocessed_essays): # for each review/sentence
# vector = np.zeros(300) # as word vectors are of zero length
# tf_idf_weight =0; # num of words with a valid vector in the sentence/review
# for word in sentence.split(): # for each word in a review/sentence
# if (word in glove_words) and (word in tfidf_words):
# vec = model[word] # getting the vector for each word
# # here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sente
# tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for e
# vector += (vec * tf_idf) # calculating tfidf weighted w2v
# tf_idf_weight += tf_idf
# if tf_idf_weight != 0:
# vector /= tf_idf_weight
# tfidf_w2v_vectors.append(vector)

# print(len(tfidf_w2v_vectors))
# print(len(tfidf_w2v_vectors[0]))

In [238]: # Similarly you can vectorize for title also

1.5.3 Vectorizing Numerical features

In [239]: price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()

project_data = pd.merge(project_data, price_data, on='id', how='left')
In [240]: # check this one: https://www.youtube.com/watch?v=0HOqOcln3Z4&t=530s
# standardization sklearn: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.ht
# from sklearn.preprocessing import StandardScaler

# # price_standardized = standardScalar.fit(project_data['price'].values)
# # this will rise the error
# # ValueError: Expected 2D array, got 1D array instead: array=[725.05 213.03 329. ... 399. 287.73 5.5 ].
# # Reshape your data either using array.reshape(-1, 1)

# price_scalar = StandardScaler()
# price_scalar.fit(project_data['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this dat
# print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")

# # Now standardize the data with above maen and variance.

# price_standardized = price_scalar.transform(project_data['price'].values.reshape(-1, 1))

In [241]: # price_standardized

1.5.4 Merging all the above features

we need to merge all the numerical vectors i.e catogorical, text, numerical vectors

11 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [242]: # print(categories_one_hot.shape)
# print(sub_categories_one_hot.shape)
# print(text_bow.shape)
# print(price_standardized.shape)
In [243]: # merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
# from scipy.sparse import hstack
# # with the same hstack function we are concatinating a sparse matrix and a dense matirx :)
# X = hstack((categories_one_hot, sub_categories_one_hot, text_bow, price_standardized))
# X.shape
In [244]: # please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label

Assignment 4: Naive Bayes

12 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

1. Apply Multinomial NaiveBayes on these feature sets

Set 1: categorical, numerical features + project_title(BOW) + preprocessed_eassay (BOW)

Set 2: categorical, numerical features + project_title(TFIDF)+ preprocessed_eassay (TFIDF)

2. The hyper paramter tuning(find best Alpha)

Find the best hyper parameter which will give the maximum AUC (https://www.appliedaicourse.com/course/applied-ai-course-online
/lessons/receiver-operating-characteristic-curve-roc-curve-and-auc-1/) value
Consider a wide range of alpha values for hyperparameter tuning, start as low as 0.00001
Find the best hyper paramter using k-fold cross validation or simple cross validation data
Use gridsearch cv or randomsearch cv or you can also write your own for loops to do this task of hyperparameter tuning

3. Feature importance

Find the top 10 features of positive class and top 10 features of negative class for both feature sets Set 1 and Set 2 using values of
`feature_log_prob_` parameter of MultinomialNB (https://scikit-learn.org/stable/modules/generated
/sklearn.naive_bayes.MultinomialNB.html) and print their corresponding feature names

4. Representation of results

You need to plot the performance of model both on train data and cross validation data for each hyper parameter, like shown in the figure.
Here on X-axis you will have alpha values, since they have a wide range, just to represent those alpha values on the graph, apply log
function on those alpha values.

Once after you found the best hyper parameter, you need to train your model with it, and find the AUC on test data and plot the ROC curve
on both train and test.

Along with plotting ROC curve, you need to print the confusion matrix (https://www.appliedaicourse.com/course/applied-ai-course-online
/lessons/confusion-matrix-tpr-fpr-fnr-tnr-1/) with predicted and original labels of test data points. Please visualize your confusion matrices
using seaborn heatmaps.

(https://seaborn.pydata.org/generated/seaborn.heatmap.html)

5. Conclusion

You need to summarize the results at the end of the notebook, summarize it in the table format. To print out a table please refer to this
prettytable library link (http://zetcode.com/python/prettytable/)

13 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

2. Naive Bayes

2.1 Splitting data into Train and cross validation(or test): Stratified Sampling
In [245]: # some preprocessing

#clean teacher prefix - keep alphabets only

def removedot(str1) :
return str1.replace('\.','',regex=True)
project_data['teacher_prefix_cleaned'] = (project_data['teacher_prefix'].fillna(" ").
replace('\.',' ',regex=True))
project_data.drop(['teacher_prefix'], axis=1, inplace=True)

# combine clean essay,title in project

project_data['prep_essay'] = preprocessed_essays
project_data['prep_title'] = preprocessed_title

del preprocessed_essays, preprocessed_title

gc.collect();
In [262]: # train test split
# from sklearn.model_selection import train_test_split

y = project_data['project_is_approved'].values #target
project_data.drop(['project_is_approved'], axis=1, inplace=True) # drop target column from project data

X = project_data

#split data into train and test with 2/3rd as train and 1/3rd as test
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, stratify=y)

#split train data into train and cv with 2/3rd as train and 1/3rd as cv
X_train, X_cv, y_train, y_cv = train_test_split(X_train, y_train, test_size=0.33, stratify=y_train)
In [263]: print(X_train['teacher_prefix_cleaned'].nunique())
print(X_cv['teacher_prefix_cleaned'].nunique())
print(X_test['teacher_prefix_cleaned'].nunique())
6
6
6

2.2 Make Data Model Ready: encoding numerical, categorical features

In [264]: project_data.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 109248 entries, 0 to 109247
Data columns (total 21 columns):
Unnamed: 0 109248 non-null int64
id 109248 non-null object
teacher_id 109248 non-null object
school_state 109248 non-null object
project_submitted_datetime 109248 non-null object
project_grade_category 109248 non-null object
project_title 109248 non-null object
project_essay_1 109248 non-null object
project_essay_2 109248 non-null object
project_essay_3 3758 non-null object
project_essay_4 3758 non-null object
project_resource_summary 109248 non-null object
teacher_number_of_previously_posted_projects 109248 non-null int64
clean_categories 109248 non-null object
clean_subcategories 109248 non-null object
essay 109248 non-null object
price 109248 non-null float64
quantity 109248 non-null int64
teacher_prefix_cleaned 109248 non-null object
prep_essay 109248 non-null object
prep_title 109248 non-null object
dtypes: float64(1), int64(3), object(17)
memory usage: 18.3+ MB

14 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [265]: print('Train,cv & test shape')

print('Train shape : ',X_train.shape,y_train.shape)
print('CV shape : ',X_cv.shape,y_cv.shape)
print('Test shape : ',X_test.shape,y_test.shape)
Train,cv & test shape
Train shape : (49041, 21) (49041,)
CV shape : (24155, 21) (24155,)
Test shape : (36052, 21) (36052,)

One-hot encoding of numerical features

In [266]: from sklearn.preprocessing import Normalizer

# function to normalize numerical features
def normalize_numeric(num_train,num_cv,num_test):

#initiate normalizer
norm = Normalizer()

#fit on train values

norm.fit(num_train.values.reshape(-1,1))

#transform train,cv & test values

num_train_norm = norm.transform(num_train.values.reshape(-1,1))
num_cv_norm = norm.transform(num_cv.values.reshape(-1,1))
num_test_norm = norm.transform(num_test.values.reshape(-1,1))

#return normalize values

# return num_train_norm,num_cv_norm,num_test_norm
return num_train_norm,num_cv_norm,num_test_norm
# normalize teacher_number_of_previously_posted_projects
n1_train_norm,n1_cv_norm,n1_test_norm = (normalize_numeric(X_train['teacher_number_of_previously_posted_projects'
X_cv['teacher_number_of_previously_posted_projects'],
X_test['teacher_number_of_previously_posted_projects'

# normalize price
n2_train_norm,n2_cv_norm,n2_test_norm = normalize_numeric(X_train['price'],X_cv['price'],X_test['price'])

# normalize quantity
n3_train_norm,n3_cv_norm,n3_test_norm = normalize_numeric(X_train['quantity'],X_cv['quantity'],X_test['quantity'

#combine numerical features and create train,cv & test for numerical encoded featuers
X_num_tr = np.hstack((n1_train_norm,n2_train_norm,n3_train_norm))
X_num_cv = np.hstack((n1_cv_norm,n2_cv_norm,n3_cv_norm))
X_num_test = np.hstack((n1_test_norm,n2_test_norm,n3_test_norm))

print('X_num_tr shape : ',X_num_tr.shape)

print('X_num_cv shape : ',X_num_cv.shape)
print('X_num_test shape : ',X_num_test.shape)
X_num_tr shape : (49041, 3)
X_num_cv shape : (24155, 3)
X_num_test shape : (36052, 3)

15 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [267]: # one-hot encoding of categorical featuers

from sklearn.feature_extraction.text import CountVectorizer
from collections import Counter
from scipy.sparse import hstack
# function to encode categorical variables
def encode_categorical(cat_train,cat_cv,cat_test):
my_counter = Counter()
for word in cat_train:
my_counter.update(word.split())

cat_dict = dict(my_counter)
sorted_cat_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))

vectorizer = CountVectorizer(vocabulary=list(sorted_cat_dict.keys()), lowercase=False, binary=True)

vectorizer.fit(cat_train.values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector

cat_train_ohe = vectorizer.transform(cat_train.values)
cat_cv_ohe = vectorizer.transform(cat_cv.values)
cat_test_ohe = vectorizer.transform(cat_test.values)

#return encoded featuers

return cat_train_ohe,cat_cv_ohe,cat_test_ohe
#return cat_train_ohe,cat_test_ohe

#one-hot encode teacher_prefix

c1_tr_ohe = pd.get_dummies(X_train['teacher_prefix_cleaned']).values
c1_cv_ohe = pd.get_dummies(X_cv['teacher_prefix_cleaned']).values
c1_test_ohe = pd.get_dummies(X_test['teacher_prefix_cleaned']).values

print('\nShape of teacher prefix cleaned')

print('Train:',c1_tr_ohe.shape)
print('CV:',c1_cv_ohe.shape)
print('Test:',c1_test_ohe.shape)

#one-hot encode school_state

c2_tr_ohe = pd.get_dummies(X_train['school_state']).values
c2_cv_ohe = pd.get_dummies(X_cv['school_state']).values
c2_test_ohe = pd.get_dummies(X_test['school_state']).values

print('\nShape of school_state')
print('Train:',c2_tr_ohe.shape)
print('CV:',c2_cv_ohe.shape)
print('Test:',c2_test_ohe.shape)

#one-hot encode project_grade_category

c3_tr_ohe = pd.get_dummies(X_train['project_grade_category']).values
c3_cv_ohe = pd.get_dummies(X_cv['project_grade_category']).values
c3_test_ohe = pd.get_dummies(X_test['project_grade_category']).values

print('\nShape of project_grade_category')
print('Train:',c3_tr_ohe.shape)
print('CV:',c3_cv_ohe.shape)
print('Test:',c3_test_ohe.shape)

#one-hot encode clean_categories

c4_tr_ohe,c4_cv_ohe,c4_test_ohe = (encode_categorical(X_train['clean_categories'],
X_cv['clean_categories'],
X_test['clean_categories']))

print('\nShape of clean_categories')
print('Train:',c4_tr_ohe.shape)
print('CV:',c4_cv_ohe.shape)
print('Test:',c4_test_ohe.shape)

#one-hot encode clean_subcategories

c5_tr_ohe,c5_cv_ohe,c5_test_ohe = (encode_categorical(X_train['clean_subcategories'],
X_cv['clean_subcategories'],
X_test['clean_subcategories']))

print('\nShape of clean_subcategories')
print('Train:',c5_tr_ohe.shape)
print('CV:',c5_cv_ohe.shape)
print('Test:',c5_test_ohe.shape)

16 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

2.3 Make Data Model Ready: encoding eassay, and project_title

In [268]: # encoding text features
from scipy.sparse import hstack
#bag of words
def bow_text(text_train,text_cv,text_test):

vectorizer = CountVectorizer(min_df=10,ngram_range=(1,4), max_features=5000)

vectorizer.fit(text_train.values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector

text_train_bow = vectorizer.transform(text_train.values)
text_cv_bow = vectorizer.transform(text_cv.values)
text_test_bow = vectorizer.transform(text_test.values)

# return encoded values

return text_train_bow,text_cv_bow,text_test_bow,vectorizer

#bow essay
essay_tr_bow,essay_cv_bow,essay_test_bow,bow_vect1 = bow_text(X_train['prep_essay'],X_cv['prep_essay'],X_test

#bow title
title_tr_bow,title_cv_bow,title_test_bow,bow_vect2 = bow_text(X_train['prep_title'],X_cv['prep_title'],X_test

#combine bow
X_bow_tr = hstack((title_tr_bow,essay_tr_bow))
X_bow_cv = hstack((title_cv_bow,essay_cv_bow))
X_bow_test = hstack((title_test_bow,essay_test_bow))

print('X_bow_tr shape : ',X_bow_tr.shape)

print('X_bow_cv shape : ',X_bow_cv.shape)
print('X_bow_test shape : ',X_bow_test.shape)

X_bow_tr shape : (49041, 9078)

X_bow_cv shape : (24155, 9078)
X_bow_test shape : (36052, 9078)

In [269]: type(bow_vect1.get_feature_names())

Out[269]: list

In [270]: from sklearn.feature_extraction.text import TfidfVectorizer

#function for tfidf vectorizer

def tfidf_text(text_train,text_cv,text_test,min_df=1):
vectorizer = TfidfVectorizer(min_df=min_df,ngram_range=(1,4),max_features=5000)
vectorizer.fit(text_train.values) # fit has to happen only on train data

# we use the fitted CountVectorizer to convert the text to vector

text_train_tfidf = vectorizer.transform(text_train.values)
text_cv_tfidf = vectorizer.transform(text_cv.values)
text_test_tfidf = vectorizer.transform(text_test.values)

# return encoded values

return text_train_tfidf,text_cv_tfidf,text_test_tfidf,vectorizer

# tfidf essay
essay_tr_tfidf,essay_cv_tfidf,essay_test_tfidf,tfidf_vect1 = tfidf_text(X_train['prep_essay'],X_cv['prep_essay'
X_test['prep_essay'],min_df=10)

# tfidf title
title_tr_tfidf,title_cv_tfidf,title_test_tfidf,tfidf_vect2 = (tfidf_text(X_train['prep_title'],X_cv['prep_title'
X_test['prep_title']))

#combine tfidf
X_tfidf_tr = hstack((title_tr_tfidf,essay_tr_tfidf))
X_tfidf_cv = hstack((title_cv_tfidf,essay_cv_tfidf))
X_tfidf_test = hstack((title_test_tfidf,essay_test_tfidf))

print('X_tfidf_tr shape : ',X_tfidf_tr.shape)

print('X_tfidf_cv shape : ',X_tfidf_cv.shape)
print('X_tfidf_test shape : ',X_tfidf_test.shape)
X_tfidf_tr shape : (49041, 10000)
X_tfidf_cv shape : (24155, 10000)
X_tfidf_test shape : (36052, 10000)

17 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

2.4 Appling NB() on different kind of featurization as mentioned in the instructions

Apply Naive Bayes on different kind of featurization as mentioned in the instructions

For Every model that you work on make sure you do the step 2 and step 3 of instrucations

2.4.1 Applying Naive Bayes on BOW, SET 1

In [271]: # function to plot confusion matrix

def plot_cm(clf,data,target,ds):
y_pred = clf.predict(data)
#confusion matrix
cm = confusion_matrix(target,y_pred)
title = "Confusion Matrix - " + ds + "\n"

tn, fp, fn, tp = cm.ravel()

# https://stackoverflow.com/questions/35572000/how-can-i-plot-a-confusion-matrix
# Code for drawing seaborn heatmaps
class_names = ['0','1']
df_heatmap = pd.DataFrame(cm, index=class_names, columns=class_names )
fig = plt.figure(figsize=(5,3))
heatmap = sns.heatmap(df_heatmap, annot=True, fmt="d")

# Setting tick labels for heatmap

heatmap.yaxis.set_ticklabels(heatmap.yaxis.get_ticklabels(), rotation=0, ha='right', fontsize=14)
heatmap.xaxis.set_ticklabels(heatmap.xaxis.get_ticklabels(), rotation=0, ha='right', fontsize=14)
plt.ylabel('True label',size=18)
plt.xlabel('Predict label',size=18)
plt.title(title,size=24)
plt.show()
In [283]: # list of numerical and categorical features
num_feat = ['teacher_number_of_previously_posted_projects','price','quantity']
cat_feat =['teacher_prefix_cleaned', 'school_state', 'project_grade_category', 'clean_categories',
'clean_subcategories']

18 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [284]: # Please write all the code with proper documentation

#combine numerical,categorical and BOW features

X_tr =hstack((X_num_tr,X_cat_tr,X_bow_tr)).tocsr()
X_cv = hstack((X_num_cv,X_cat_cv,X_bow_cv)).tocsr()
X_te = hstack((X_num_test,X_cat_test,X_bow_test)).tocsr()

# bow total feature list

bow_total_feat = num_feat + cat_feat + bow_vect2.get_feature_names() + bow_vect1.get_feature_names()

print('Final data matrix')

print(X_tr.shape,y_train.shape)
print(X_cv.shape,y_cv.shape)
print(X_te.shape,y_test.shape)

import matplotlib.pyplot as plt

from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import roc_auc_score

train_auc = []
cv_auc = []
a = [1000,100,10,1, 0.1, 0.01, 0.001, 0.0001,0.00001]
for i in a:
clf = MultinomialNB(alpha=i)
clf.fit(X_tr, y_train)

y_train_pred = clf.predict_proba(X_tr)
y_cv_pred = clf.predict_proba(X_cv)

# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs
train_auc.append(roc_auc_score(y_train,y_train_pred[:,1]))
cv_auc.append(roc_auc_score(y_cv, y_cv_pred[:,1]))

plt.plot(np.log(a), train_auc, label='Train AUC')

plt.plot(np.log(a), cv_auc, label='CV AUC')

plt.scatter(np.log(a), train_auc, label='Train AUC points')

plt.scatter(np.log(a), cv_auc, label='CV AUC points')

plt.legend()
plt.xlabel("alpha: hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()
Final data matrix
(49041, 9181) (49041,)
(24155, 9181) (24155,)
(36052, 9181) (36052,)

In [285]: a=1

19 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [286]: from sklearn.metrics import roc_curve, auc, confusion_matrix

clf = MultinomialNB(alpha=a)
clf.fit(X_tr, y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs

y_train_pred = clf.predict_proba(X_tr)
y_test_pred = clf.predict_proba(X_te)

train_fpr, train_tpr, tr_thresholds = roc_curve(y_train, y_train_pred[:,1])

test_fpr, test_tpr, te_thresholds = roc_curve(y_test, y_test_pred[:,1])

plt.plot(train_fpr, train_tpr, label="train AUC ="+str(auc(train_fpr, train_tpr)))

plt.plot(test_fpr, test_tpr, label="test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.xlabel("alpha: hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()

#print('Confusion matrix - Train')

plot_cm(clf,X_tr,y_train,"Train")

#print('\n\nConfusion matrix - Test')

plot_cm(clf,X_te,y_test,"Test")

20 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

2.4.1.1 Top 10 important features of positive class from SET 1

In [287]: # Please write all the code with proper documentation

#https://stackoverflow.com/questions/50526898/how-to-get-feature-importance-in-naive-bayes
pos_class_prob_sorted = clf.feature_log_prob_[1, :].argsort()
print(np.take(bow_total_feat, pos_class_prob_sorted[:10]))
['teacher_prefix_cleaned' 'school_state' 'reading is fun'
'the write stuff' 'plants' 'it move' 'softball' 'successful students'
'idea' 'library needs']

2.4.1.2 Top 10 important features of negative class from SET 1

In [288]: # Please write all the code with proper documentation

# https://stackoverflow.com/questions/50526898/how-to-get-feature-importance-in-naive-bayes
neg_class_prob_sorted = clf.feature_log_prob_[0, :].argsort()
print(np.take(bow_total_feat, neg_class_prob_sorted[:10]))
['like move it move' 'up with good book' 'bees' 'languages' 'new teacher'
'with new' 'entrepreneurs' 'reading' 'carpet' 'readers writers']

2.4.2 Applying Naive Bayes on TFIDF, SET 2

21 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [289]: # Please write all the code with proper documentation

#combine numerical,categorical and tfidf features

X_tr1 =hstack((X_num_tr,X_cat_tr,X_tfidf_tr)).tocsr()
X_cv1 = hstack((X_num_cv,X_cat_cv,X_tfidf_cv)).tocsr()
X_te1 = hstack((X_num_test,X_cat_test,X_tfidf_test)).tocsr()

# tfidf total feature list

tfidf_total_feat = num_feat + cat_feat + tfidf_vect2.get_feature_names() + tfidf_vect1.get_feature_names()

print('Final data matrix')

print(X_tr1.shape,y_train.shape)
print(X_cv1.shape,y_cv.shape)
print(X_te1.shape,y_test.shape)

train_auc = []
cv_auc = []
a = [1000,100,10,1, 0.1, 0.01, 0.001, 0.0001,0.00001]
for i in a:
clf = MultinomialNB(alpha=i)
clf.fit(X_tr1, y_train)

y_train_pred = clf.predict_proba(X_tr1)
y_cv_pred = clf.predict_proba(X_cv1)

# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs
train_auc.append(roc_auc_score(y_train,y_train_pred[:,1]))
cv_auc.append(roc_auc_score(y_cv, y_cv_pred[:,1]))

plt.plot(np.log(a), train_auc, label='Train AUC')

plt.plot(np.log(a), cv_auc, label='CV AUC')

plt.scatter(np.log(a), train_auc, label='Train AUC points')

plt.scatter(np.log(a), cv_auc, label='CV AUC points')

plt.legend()
plt.xlabel("alpha: hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()
Final data matrix
(49041, 10103) (49041,)
(24155, 10103) (24155,)
(36052, 10103) (36052,)

In [301]: a=1

22 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [302]: clf = MultinomialNB(alpha=a)

clf.fit(X_tr1, y_train)
# roc_auc_score(y_true, y_score) the 2nd parameter should be probability estimates of the positive class
# not the predicted outputs

y_train_pred = clf.predict_proba(X_tr1)
y_test_pred = clf.predict_proba(X_te1)

train_fpr, train_tpr, tr_thresholds = roc_curve(y_train, y_train_pred[:,1])

test_fpr, test_tpr, te_thresholds = roc_curve(y_test, y_test_pred[:,1])

plt.plot(train_fpr, train_tpr, label="train AUC ="+str(auc(train_fpr, train_tpr)))

plt.plot(test_fpr, test_tpr, label="test AUC ="+str(auc(test_fpr, test_tpr)))
plt.legend()
plt.xlabel("alpha: hyperparameter")
plt.ylabel("AUC")
plt.title("ERROR PLOTS")
plt.grid()
plt.show()

#print('Confusion matrix - Train')

plot_cm(clf,X_tr1,y_train,"Train")

#print('\n\nConfusion matrix - Test')

plot_cm(clf,X_te1,y_test,"Test")

2.4.2.1 Top 10 important features of positive class from SET 2

23 of 24 4/17/2019, 6:36 PM
4_DonorsChoose_NB http://localhost:8888/notebooks/Google Drive/appliedai/assignments/Do...

In [303]: # Please write all the code with proper documentation

# Please write all the code with proper documentation
pos_class_prob_sorted = clf.feature_log_prob_[1, :].argsort()
print(np.take(bow_total_feat, pos_class_prob_sorted[:10]))
['teacher_prefix_cleaned' 'school_state' 'classroom nannan' 'cutting'
'resource' 'learning through play' 'researching' 'class time'
'bag chairs' 'bags']

2.4.2.2 Top 10 important features of negative class from SET 2

In [304]: # Please write all the code with proper documentation

neg_class_prob_sorted = clf.feature_log_prob_[0, :].argsort()
print(np.take(bow_total_feat, neg_class_prob_sorted[:10]))
['mo' 'picture books' 'close' 'pick' 'co' 'coach' 'performing' 'perfect'
'collaboration' 'park']

3. Conclusions
In [307]: # Please compare all your models using Prettytable library
from prettytable import PrettyTable
x = PrettyTable()
x.field_names = ["Vectorize","Model","HyperParam(alpha)","AUC"]
x.add_row(['BOW','MultinomialNB',1,.695])
x.add_row(['TFIDF','MultinomialNB',1,0.662])
print(x)
+-----------+---------------+-------------------+-------+
| Vectorize | Model | HyperParam(alpha) | AUC |
+-----------+---------------+-------------------+-------+
| BOW | MultinomialNB | 1 | 0.695 |
| TFIDF | MultinomialNB | 1 | 0.662 |
+-----------+---------------+-------------------+-------+

1. Multinomial NB with Bag Of Words performs better than with TF-IDF.

2. Multinomial NB in general performs better and faster than kNN models.
3. Two categorical features teacher_prefix_cleaned & school_state are most important for predicting positive class!
4. Multinomial NB with TFIDF features peforms worse than as compared to with Multinomial NB with BOW features for predicting negative class
correctly.

24 of 24 4/17/2019, 6:36 PM