Matrices
Professor’s name:
Student’s name:
Date:
The terms positive outcome and negative are used during the discussion of confusion matrices. This translates to the fact that individuals are mandated to understand the existing differences and the cost implications of both types I and type II errors. Various statistical concepts are established as being confusing, which are at times established as being counter-intuitive. Thus, it is, at times, ordeal with subject to establishing the explanation of differences with regards to the null hypothesis and alternative hypothesis (Bramer, 2016).
The type I error, which is established as being false positive, is established as being a result indicating the presence of a certain condition in a situation whereby it is actually not present. On the other side, the type II error established as the false negative is established; as a result indicating the absence of a certain condition in a situation whereby the condition is present (Pustejovsky & Stubbs, 2012).
For instance, assuming that I am a shepherd who is characterized by bad eyesight subject to difficulties in making distinctions between a sheepdog and a wolf. This is established as being a bad trait with subject to the fact that the cost implication of being wrong is established as being expensive whereby there is a situation whereby I would make a confusion regarding a wolf being a sheepdog meaning I take no action that would cost me $2000 in every single occurrence that has a probability of 10% chance of occurrence. On the other hand, there are situations I would make a confusion regarding the sheep wolf being the dog whereby I end up accidentally killing the sheepdog consequently jeopardizing the security of my flock which would cost me approximately $5000 having an occurrence chance of 5%. From this illustration, we can assign a false positive in the first instance since its incorrect to make an identification of the animal as being the wolf while in reality, it is a sheepdog. On the second instance, we can assign a false negative based on the fact that I would have made an incorrect identification subject to identifying the animal as the sheepdog when actually it’s the wolf (Keller, 2
References
Bramer, M. (2016). Principles of data mining. Springer.
Keller, T. (2011). Mining the Internet of things: Detection of false-positive RFID tag reads using low-level reader data.
Pustejovsky, J., & Stubbs, A. (2012). Natural language annotation for machine learning: A guide to corpus-building for applications. O’Reilly Media.
011).