Hands-On Machine Learning for Algorithmic Trading

Building a dataset of restaurant bookings

Now you only need to combine all the interesting elements from the website to create a feature that you could use in a model to predict economic activity in geographic regions or foot traffic in specific neighborhoods.

With Selenium, you can follow the links to the next pages and quickly build a dataset of over 10,000 restaurants in NYC that you could then update periodically to track a time series. First, we set up a function that parses the content of the pages that we plan on crawling:

def parse_html(html):    data, item = pd.DataFrame(), {}    soup = BeautifulSoup(html, 'lxml')    for i, resto in enumerate(soup.find_all('div', class_='rest-row-           info')):        item['name'] = resto.find('span', class_='rest-row-name- ...