**Augmenting Naive Bayes Classifiers with Statistical Language Models**

We augment naive Thomas Bayes|mathematician} models with applied mathematics n-gram language models to deal with short-comings of the quality naive Bayes text classifier. The result’s a generalized naive Thomas Bayes|mathematician} classifier that permits for an area Markoff dependence among observations; a model we have a tendency to talk over with because the ChainAugmentedNaiveBayes (CAN) Bayes classifier. will models have 2 blessings over customary naive Thomas Bayes classifiers. First, they relax a number of the independence assumptions of naive Bayes—allowing an area Markoff chain dependence within the ascertained variables—while still allowing economical reasoning and learning. Second, they enable easy application of refined smoothing techniques from applied mathematics language modeling, that permits one to get higher parameter estimates than the quality Marquis de Laplace smoothing utilized in naive Thomas Bayes classification. during this paper, we have a tendency to introduce will models and apply them to varied text classification issues. **[1]**

**Intrusion Detection using Naive Bayes Classifier with Feature Reduction**

Intrusion discoverion is that the method of observation and analyzing the events occurring in a very ADPS so as to detect signs of security issues. these days most of the intrusion detection approaches targeted on the problems of feature choice or reduction, since a number of the options are extraneous Associate in Nursingd redundant which ends prolonged detection method and degrades the performance of an intrusion detection system (IDS). the aim of this study is to spot vital reduced input options in building IDS that’s computationally economical and effective. For this we tend to investigate the performance of 3 commonplace feature choice strategies victimisation Correlation-based Feature choice, info Gain and Gain magnitude relation. during this paper we tend to propose technique Feature Vitality primarily based Reduction technique, to spot vital reduced input options. **[2]**

**Privacy Preserving Naïve Bayes Classifier for Vertically Partitioned Data**

Privacy-Preserving data processing – developing models while not seeing the info – is receiving growing attention. This paper assumes a privacy-preserving distributed data processing scenario: knowledge sources collaborate to develop a worldwide model, however should not disclose their knowledge to others. Naïve Thomas Bayes is usually used as a baseline classifier, systematically providing cheap classification performance. This paper brings privacy-preservation to Naïve Thomas Bayes classification on vertically divided knowledge. **[3]**

**Employing fingerprinting of medicinal plants by means of LC-MS and machine learning for species identification task**

A dataset of liquid chromatography-mass spectrum analysis measurements of healthful plant extracts from seventy four species was generated and used for coaching and verificatory plant species identification algorithms. numerous methods for knowledge handling and have house extraction were tested. unnatural Tucker decomposition, large-scale (more than 1500 variables) separate theorem Networks and autoencoder based mostly spatial property reduction plus continuous Thomas Bayes classifier and supply regression were optimized to attain the simplest accuracy. Even with elimination of all retention time values accuracies of up to ninety six and ninety two were achieved on validation set for plant species and plant part identification severally.** [4]**

**Hypertension Prediction System Using Naive Bayes Classifier**

Hypertension is Associate in Nursing malady that usually ends up in severe and severe diseases like failure, arteria sickness, heart failure and different severe conditions if not promptly diagnosed and treated. data processing the employment of a spread of techniques to smoothen data discovery or decision-making data within the information and extracting these during a manner that they will place to use in areas like predictions, prediction and estimation. This analysis has developed pressure level|cardiovascular disease} prognosticative system victimisation data processing modelling technique, namely, Naïve Thomas Bayes. Medical profiles like age, sex, blood pressure, pain and glucose it will predict the probability of patients obtaining a high blood pressure. This work was enforced in maori hen setting as Associate in Nursing application that takes medical test’s parameter as an input. **[5]**

**Reference**

**[1]** Peng, F., Schuurmans, D. and Wang, S., 2004. Augmenting naive bayes classifiers with statistical language models. Information Retrieval, 7(3-4), (Web Link)

**[2]** Mukherjee, S. and Sharma, N., 2012. Intrusion detection using naive Bayes classifier with feature reduction. Procedia Technology, 4, (Web Link)

**[3]** Vaidya, J. and Clifton, C., 2004, April. Privacy preserving naive bayes classifier for vertically partitioned data. In Proceedings of the 2004 SIAM International Conference on Data Mining (pp. 522-526). Society for Industrial and Applied Mathematics. (Web Link)

**[4]** Employing fingerprinting of medicinal plants by means of LC-MS and machine learning for species identification task

Pavel Kharyuk, Dmitry Nazarenko, Ivan Oseledets, Igor Rodin, Oleg Shpigun, Andrey Tsitsilin & Mikhail Lavrentyev

Scientific Reports volume 8, Article number: 17053 (2018) (Web Link)

**[5]** Afeni, B. O., Aruleba, T. I. and Oloyede, I. A. (2017) “Hypertension Prediction System Using Naive Bayes Classifier”, Journal of Advances in Mathematics and Computer Science, 24(2), (Web Link)