Comprehensive Response Describing Naive Bayes Essay

Comprehensive Response Describing Naive Bayes Essay

Question one

Naive Bayes is a type of algorithm that is used in the classification of emails through detecting spam. .The algorithm classifies data with the assumptions that the data is independent and there are no interconnections of the features present in one feature to other (Tang, Kay & He, 2016). This type of algorithm is based on probability application that helps in fixing the missing information from the model statistics. Naive Bayes classification is used in big data of text mining and in the classifying of different documents. Naive Bayes model of classification is simple and quick to use even if the data given is large. The classification algorithm can also be used in complexity classification (Zhang et al., 2016). The demerit of naive Bayer is that it classifies data quicker because the classification is based on the assumptions thus makings it reliable for big data classification than the discriminative algorithms such as logistic and linear regression algorithms. The Naive Bayes algorithm attempts to fit the individual wants and detects any presence of false positive spam in the texts.

BUY A PLAGIARISM-FREE PAPER HERE

Question two.

In classification, naive Bayes filters spam by finding the connections of different variables that depend on each other. This is basically done through finding the correlation of email with spam and those don’t have. Once the associations of the variables have been identified the Bayes theorem is applied to calculate the probability whether the spam is present in the email or not (Feng et al., 2016).The probabilities are then used in the description of the conditions of the features present in the text and how associated with the other feature. In the filtering process, the naive classifier uses the idea that different words have a distinguished event of occurrence in email. The filter is firstly trained on probabilities, this involves manually indication by the user whether the new email received has spam or not, this enables the filter probability to be build upon and thus can automatically detect if spam is present is in the new email of the user’s inbox. Once the filter has been trained, the probabilities of each word are used to calculate the probability whether particular words in the email are spam or not (Shafi’I et al., 2017). During computation of probabilities of the words in the email, when detections move beyond a given threshold the email is considered spam. The spam email can either be deleted or stored to the “junk” after filtering. Comprehensive Response Describing Naive Bayes Essay

Question three.

Naive Bayes classification is highly deployed in insurance companies to assess the risk associated with insured customers. To improve efficiency in the classification insurance customers are insured depending on their risk levels. The risks insured by different classifications are then used to compute the probability of the risk to occur this helps in setting premiums for different clients. During membership of new clients, naive Bayes’ techniques are used to compare the records of the client’s information and calculate the membership probabilities that are used in assigning the customer to the appropriate group (Jing et al., 2016). Additionally, naive Bayes algorithm techniques have been very important in detecting false claims from the fraudsters. The techniques are used to identify fraud through computing the probabilities of the risks insured with the interrelationship of the cause of the loss.

References

Feng, W., Sun, J., Zhang, L., Cao, C., & Yang, Q. (2016, December). A support vector machine based naive Bayes algorithm for spam filtering. In 2016 IEEE 35th International Performance Computing and Communications Conference (IPCCC) (pp. 1-8). IEEE.

Jing, L., Zhao, W., Sharma, K., & Feng, R. (2018, January). Research on Probability-based Learning Application on Car Insurance Data. In 2017 4th International Conference on Machinery, Materials, and Computer (MACMC 2017). Atlantis Press.

Shafi’I, M. A., Latiff, M. S. A., Chiroma, H., Osho, O., Abdul-Salaam, G., Abubakar, A. I., & Herawan, T. (2017). A review on mobile SMS spam filtering techniques. IEEE Access5, 15650-15666.

Tang, B., Kay, S., & He, H. (2016). Toward optimal feature selection in naive Bayes for text categorization. IEEE transactions on knowledge and data engineering28(9), 2508-2521.

Zhang, L., Jiang, L., Li, C., & Kong, G. (2016). Two feature weighting approaches for naive Bayes text classifiers. Knowledge-Based Systems100, 137-144.

Comprehensive Response Describing Naive Bayes Essay