Peer responses must be substantive in nature.
Discussion 1
In statistics, the concept of a categorical attribute refers to a component in a model that can
only take on certain predetermined values. As a result, the various choices for the component do
not come together to produce a continuous gradient (as in the case of length or time). A few
examples of categorical qualities include things like colors, different kinds of forms, names of
persons or organizations, and so on.
According to Barker (1980), categorical data types are qualities that are handled as if they
were separate symbols or merely names. Because it may take on values such as black, green, blue,
gray, etc., the color of the iris of the human eye is an example of a categorical data type. Because
there is no direct link between the data values, it is not possible to apply any mathematical
operators—with the exception of the logical or “is equal” operator—to the data. It is important to
keep in mind that certain categorical qualities may sometimes be expressed as continuous or
smooth traits.
A continuous attribute is one that accepts values determined via the use of measurement.
As an example, a person’s height is a continuous dimension since it can be measured. Continuous
data are not limited to individual values that have been established, but rather may occupy any
value within a range. A person’s weight may be expressed as 90 pounds, 90.5 pounds, 90.12
pounds, or 90.345 pounds, and so on. Numeric representations are always used for continuous
data.
Continuous data provide for a little more wiggle room when it comes to accuracy and also
make it possible to interpolate (Tan, 2018). The concept of proportion is given its full significance.
You probably don’t count the quantity of molecules in your bottle of milk for each individual
ingredient that makes up milk unless you are a scientist. You are making a pastry, right? For
instance, the quantity of milk is measured in liters. If your oven is too small for the cake recipe
you want to make, it is typically OK to cut all of the ingredients in half. You won’t be able to tell
the difference between a piece of this interpolated cake and the identical piece of cake you would
have gotten if you had followed the original recipe.
References
Barker, K. N. (1980). Data Collection Techniques: Observation.
Tan, P.-N. (2018). Introduction to Data Mining. Pearson.
Discussion 2:
Question 1
There are three main techniques for handling categorical attributes namely nominal,
ordinal, and cardinal (Mougan et al., 2023). When deciding which one to use, the first thing you
should consider is whether the information in each category is equal or not. If it is not equal, you
should think about whether or not the order matters.
Question 2
Continuous attributes are those that have a numeric value, such as height, weight, or age.
On the other hand, categorical attributes are those that are named groups of things, such as
gender or ethnic group. Continuous and categorical attributes differ from each other in several
ways (Mary et al., 2019). In general, continuous variables quantify a quantity in terms of an
amount or extent of something, while categorical variables classify people according to certain
characteristics.
Question 3
A concept hierarchy is a way of organizing concepts into levels based on their relatedness
(Weng & Luo, 2023). It is sometimes called a topic tree, mind map, or taxonomy. The purpose of
a concept hierarchy is to enhance the ease with which people can find the information they need.
Question 4
There are several common patterns that include the following. Firstly, it is a linear pattern
that indicates a direct relationship between two factors, such as the amount of time spent doing
an assigned task and the accuracy of the task (Hewamalage et al., 2022). A non-linear pattern
usually indicates that there is a threshold value below which something happens reliably but
above it there is either no effect at all or an opposite effect. A cyclical pattern shows up when
two factors influence each other in cycles or cycles are caused by one factor e.g., temperature
and seasons (Hewamalage et al., 2022).
References
Hewamalage, H., Bergmeir, C., & Bandara, K. (2022). Global models for time series forecasting:
A simulation study. Pattern Recognition, 124, 108441.
Mary, J., Calauzenes, C., & El Karoui, N. (2019, May). Fairness-aware learning for continuous
attributes and treatments. In International Conference on Machine Learning (pp. 43824391). PMLR.
Mougan, C., Álvarez, J. M., Ruggieri, S., & Staab, S. (2023, August). Fairness implications of
encoding protected categorical attributes. In Proceedings of the 2023 AAAI/ACM
Conference on AI, Ethics, and Society (pp. 454-465).
Weng, W., & Luo, W. (2023). A Comparative Analysis of Data Mining Methods and
Hierarchical Linear Modeling Using PISA 2018 Data. International Journal of Database
Management Systems (IJDMS) Vol, 15.