Page 221 - AI Ver 1.0 Class 10
P. 221
How will these factors affect our problem statement? The answer to this is the System Maps tool. It provides
the relationship of various factors and their impact on the project goal. The following is the system map for the
above scenario.
+
–
Piece of
Dish Dish Qty
Prepared
Per Day
–
+
Dish
Consumption
+ + Total
Customers
+
–
Qty of Dish
Unconsumed for Next Day
Dish Qty
Per day –
The above system map shows the relationship of various data features with each other. The goal of the project is
to reduce food wastage, each element is defined to depict its impact on the goal. An arrow with a positive sign
shows a direct relation and the one with negative signs shows an inverse relationship with the elements.
For our scenario, a dataset that covers the details of all elements for making a specific dish is prepared by the
restaurant. The data is collected over the past 30 days through surveys as it caters to a specific restaurant only.
The elements are under the following heads— Quantity of dish produced per day, Quantity of unconsumed dish
each day, Price of the dish, Total footfall of the day, Number of fixed customers, etc.
Stage 3: Data Exploration
The next stage is data exploration, it is the stage of cleaning the acquired data. Once the data is collected, we
extract meaningful information from it to achieve the goal of our project. In our scenario, the goal is to develop an
effective system that can predict the quantity of food to be prepared by restaurants to minimise wastage.
The following data needs to be prepared for the same:
• Name of the dish.
• Quantity of the dish prepared every day.
• Unconsumed quantity of dish per day.
Stage 4: Data Modelling
After the data is explored and the dataset is ready, the next stage is data modelling. In this, the model is fed with
the training dataset to see if it matches the desired output. In this scenario, the Regression model is chosen.
Regression is a supervised learning model, which is used to predict continuous values. In our scenario the data
set that we shall use to train our model is a continuous data of 30 days, we can use a regression model so that
it can predict the next values. In this case, the model is trained for the first 20 days and then gets evaluated for
the rest 10 days.
Data Science 219

