
from pytrends.request import TrendReq
import datetime as dt
import matplotlib.pyplot as plt
plt.rcParams['figure.figsize'] = [30, 30]
%matplotlib inline  
import numpy as np
import pandas as pd
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error, r2_score
import seaborn as sns


pytrends = TrendReq(hl='en-US', tz=360)  # trend request
kw_list = ["\"alcohol abuser\" + \"alcohol abusers\" + \"alcohol abuse\"", "\"alcohol user\" + \"alcohol users\" + \"alcohol use\"" ]  # keyword list
pytrends.build_payload(kw_list, cat=0, timeframe='2004-01-01 2021-12-31', geo='US', gprop='')  # build pytrends data
trendinfo = pd.DataFrame(pytrends.interest_over_time())  # interest over time, dataframe format


# SHOWS ALL ROWS
#pd.set_option('display.max_rows',trendinfo.shape[0]+1) 

# SHOWS HEAD (first 5 rows in dataframe)
trendinfo.head()


indices = trendinfo.index  # in datetime format

x = np.array(trendinfo['"alcohol abuser" + "alcohol abusers" + "alcohol abuse"']).reshape((-1, 1)) # alcohol use array
y = np.array(trendinfo['"alcohol user" + "alcohol users" + "alcohol use"']).reshape((-1, 1)) # alcohol abuse array

fig, ax = plt.subplots()
ax.scatter(indices, y, color='g', s=1,label='\"alcohol use\" OR \"alcohol user\" OR \"alcohol users\"')

ax.scatter(indices, x, color='r',s=1, label='\"alcohol abuse\" OR \"alcohol abuser\" OR \"alcohol abusers\"')

plt.legend(bbox_to_anchor=(.02, 1), loc='upper left', borderaxespad=0.)
plt.ylabel('Search Interest')
plt.xlabel('Year')

plt.show()
fig.savefig("alcohol_use___alcohol_abuse.png", dpi=300)

	"alcohol abuser" + "alcohol abusers" + "alcohol abuse"	"alcohol user" + "alcohol users" + "alcohol use"	isPartial
date
2004-01-01	62	11	False
2004-02-01	64	21	False
2004-03-01	50	23	False
2004-04-01	75	34	False
2004-05-01	64	26	False

"Alcohol use" / "Alcohol abuse"¶