#!/usr/bin/env python
# coding: utf-8

# # Chapter 1: Introduction
# 
# Let's build a vector for input text, e.g., from `doc1`:

# In[14]:


doc1 = "meeting ... management ... meeting ... management ... meeting "
doc1 += "... management ... meeting ... meeting"

vector = [0, 0]

for word in doc1.split(" "):
    if word=="management":
        vector[0] = vector[0] + 1
    if word=="meeting":
        vector[1] = vector[1] + 1
        
print (vector)


# Here is how you can calculate *Euclidean distance* between a document and a query:

# In[15]:


import math

query = [1, 1]
doc1 = [3, 5]
sq_length = 0

for index in range(0, len(query)):
    sq_length += math.pow((doc1[index] - query[index]), 2)
            
print (math.sqrt(sq_length))


# Finally, let's estimate *cosine similarity*:

# In[16]:


import math

query = [1, 1]
doc1 = [3, 5]

def length(vector):
    sq_length = 0
    for index in range(0, len(vector)):
        sq_length += math.pow(vector[index], 2)
    return math.sqrt(sq_length)
    
def dot_product(vector1, vector2):
    if len(vector1)==len(vector2):
        dot_prod = 0
        for index in range(0, len(vector1)):
            dot_prod += vector1[index]*vector2[index]
        return dot_prod
    else:
        return "Unmatching dimensionality"

cosine=dot_product(query, doc1)/(length(query)*length(doc1))
print (cosine)


# In[ ]: