In data mining, we must construct a fixed-length feature vector representation f

Question

In data mining, we must construct a fixed-length feature vector representation for every observation before applying data mining methods on it. Describe three valid features that we could potentially extract from a physical fruit to predict the type of fruit and train a classifier that could correctly distinguish between fruits. For each of the examples listed below, specify which of the four data mining branches that we could solve it with. Justify your reasoning with a single sentence for each. Discovering different groups of customers Forecasting the wind speed based on previous historical collections Predicting whether a patient has a disease based on previous collected genes who have had the disease. Why does clustering not fall under the paradigm of supervised data mining? Justify your reasoning with a single sentence.

Explanation / Answer

Data Mining:

Data mining is a process aimed at finding set patterns in enormous data sets. The main purpose of data mining is to extract information from a large amount of data ( also known as a data set). The extracted data is then modified into an understandable structure.Data mining process involves database and data management aspects, and data pre-processing.

Feature Vector:

A feature vector is a vector of attributes for a database record.It contains the set of database attributes that uniquely identify each data member.

Types of Attributes:

In the specified situation, we want to look at features that specifies the type of fruit. The possible features can be any of the following options:

1) Season

2) Name

3) Number_of_seeds

4) Taste(bitter, sweet, sour)

Related Questions

Navigate

• Browse (All)
• Browse I
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.