Every day, we make decisions based on the information we have in our consciousness. We might order a cheeseburger from place X, possibly because we’ve tried it before, and this particular order meets certain criteria that lead us to make that decision. For a decision like this, we might not need (at least not explicitly) a lot of data, right? Wrong.
For a decision like this, we unconsciously evaluate at least the following variables:
- Personal Preferences: This includes the type of bread, choice of meat (beef, chicken, vegetarian), type of cheese, and preferences for condiments or sauces.
- Past Experiences: How were your previous experiences with other burgers? Did a specific place leave a good or bad impression?
- Recommendations: Opinions from friends or customer reviews can influence your choice, especially if many people recommend or criticize a specific place.
- Price: The cost of the burger can be a deciding factor, especially when comparing options from different establishments.
- Ingredients and Quality: The origin of ingredients, freshness, and overall quality of the burger are important.
- Dietary Restrictions: Considerations about food allergies, vegetarian or vegan diets, or religious restrictions may rule out certain options.
- Hunger: The level of hunger can influence the size of the burger you desire or the sides you might add, such as fries, dessert, or even a salad.
- Convenience: The proximity of the establishment, delivery options, and estimated wait time can be decisive factors.
- Emotional State: Sometimes, the craving for a certain type of burger can be influenced by your emotional state or nostalgic memories associated with a dining experience.
- Weather Conditions: On a cold day, you might prefer a place that offers quick delivery, or on a hot day, a lighter burger might be more appealing.
Bringing this same scenario to data science…
Imagine you have an advanced framework that utilizes data science, equipped with machine learning algorithms, to help you decide where to order a burger. This system collects and analyses a large volume of data related to your past burger experiences, including ingredients, establishments, taste reviews, delivery time, and even how you felt after eating the burgers (if you provided that information to the algorithm) and other factors that could be included, use your imagination.
Every time you think about ordering a burger, the system applies predictive models to evaluate all available options, considering your personal preferences and consumption history. It processes information such as: which places you liked the most in the past, which ingredients you prefer, how the weather is, and even sentimental analyses of your reactions expressed in comments or online reviews. The system can even take into account external data, such as changes in the menus of establishments or new reviews from other users.
Just as you use your consciousness and memory to choose the cheeseburger from place X, the data science system uses an analytical process based on data to arrive at the same decision. It weighs all the factors, makes predictions about what you’ll likely enjoy most, and recommends the burger based on that analysis.
The fundamental difference is that while your decision is influenced by memories and subjective sensations accumulated in your memory, the decision of the data science system is derived from an objective analytical process, which quantifies and evaluates patterns in large volumes of data to make the best possible recommendation based on historical and predictive evidence.
Final Thoughts
In a world where technology advances daily, competition increases substantially, and vast amounts of data are produced every second, using data science to make a decision (speaking of business) is an absolutely valuable tool. It can reduce the margin of error in choices, help decide which direction to take, learn what not to do before doing it, and even provide detailed planning for your business both in the short and long term.
Blog article written by:
VITOR DALLAGNOLO
Data Engineer
