MAC: A Multiclass Associative Classification Algorithm

Associative classification (AC) is a data mining approach that uses association rule discovery methods to build classification systems (classifiers). Several research studies reveal that AC normally generates higher accurate classifiers than classic classification data mining approaches such as rule induction, probabilistic and decision trees. This paper proposes a new multiclass AC algorithm called MAC. The proposed algorithm employs a novel method for building the classifier that normally reduces the resulting classifier size in order to enable end-user to more understand and maintain it. Experimentations against 19 different data sets from the UCI data repository and using different common AC and traditional learning approaches have been conducted with reference to classification accuracy and the number of rules derived. The results show that the proposed algorithm is able to derive higher predictive classifiers than rule induction (RIPPER) and decision tree (C4.5) algorithms and very competitive to a known AC algorithm named MCAR. Furthermore, MAC is also able to produce less number of rules than MCAR in normal circumstances (standard support and confidence thresholds) and in sever circumstances (low support and confidence thresholds) and for most of the data sets considered in the experiments.

ACRIPPER: A New Associative Classification Based on RIPPER Algorithm

Associative Classification (AC) classifiers are of substantial interest due to their ability to be utilised for mining vast sets of rules. However, researchers over the decades have shown that a large number of these mined rules are trivial, irrelevant, redundant, and sometimes harmful, as they can cause decision-making bias. Accordingly, in our paper, we address these challenges and propose a new novel AC approach based on the RIPPER algorithm, which we refer to as ACRIPPER. Our new approach combines the strength of the RIPPER algorithm with the classical AC method, in order to achieve: (1) a reduction in the number of rules being mined, especially those rules that are largely insignificant; (2) a high level of integration among the confidence and support of the rules on one hand and the class imbalance level in the prediction phase on the other hand. Our experimental results, using 20 different well-known datasets, reveal that the proposed ACRIPPER significantly outperforms the well-known rule-based algorithms RIPPER and J48. Moreover, ACRIPPER significantly outperforms the current AC-based algorithms CBA, CMAR, ECBA, FACA, and ACPRISM. Finally, ACRIPPER is found to achieve the best average and ranking on the accuracy measure.

Stuttering Disfluency Detection Using Machine Learning Approaches

Stuttering is a neurodevelopmental speech disorder wherein people suffer from disfluency in speech generation. Recent research has applied machine learning and deep learning approaches to stuttering disfluency recognition and classification. However, these studies have focussed on small datasets, generated by a limited number of speakers and within specific tasks, such as reading. This paper rigorously investigates the effective use of eight well-known machine learning classifiers, on two publicly available datasets (FluencyBank and SEP-28k) to automatically detect stuttering disfluency using multiple objective metrics, i.e. prediction accuracy, recall, precision, F1-score, and AUC measures. Our experimental results on the two datasets show that the Random Forest classifier achieves the best performance, with an accuracy of 50.3% and 50.35%, a recall of 50% and 42%, a precision of 42% and 46%, and an F1 score of 42% and 34%, against the FluencyBank and SEP-28K datasets, respectively. Moreover, we show that the machine learning-based approaches may not be effective in accurate stuttering disfluency evaluation, due to diverse variations in speech rate, and differences in vocal tracts between children and adults. We argue that the use of deep learning approaches and Automatic Speech Recognition (ASR) with language models may improve outcomes, specifically for large scale and imbalanced datasets.

Isolation and identification of marine Bacillus altitudinis KB1 from coastal Kerala: asparaginase producer

L-asparaginase is a target for many researchers as its properties against cancer, especially leukaemia, and protective agents reduce acrylamide in fried food. In this study, the water samples from Thumba Arattuvazhi Beach in Kerala were screened for l-asparaginase producing microorganisms. This was followed by colourimetric screening using modified M9 media with 0.009% Phenol red dye and using l-asparagine as a sole nitrogen source. Then, the Nessler assay was performed to quantify the enzyme. Molecular identification was made by 16SrRNA sequencing and aligned the sequence with GeneBank for phylogenetic tree construction using BLAST. Seawater was serially diluted for 10-1 to 10-6 using nutrient agar plates. A total of 19 bacterial colonies were isolated. The colonies were evaluated to produce l-asparaginase according to the pink zone around the colonies on the modified M9 medium using a red phenol indicator. The KB1 sample was selected for further studies according to plate colour assay. Nessler assay of L-asparaginase quantified as 2.537 IU/ml. Molecular characterisation showed the sequence association with Bacillus altitudinis the sequence submitted in Genebank as B. altitudinis KB1 strain. The l-asparaginase II gene (AnsB) was amplified based on the entire length of the hypothetical protein of annotated genome with accession number CP022319.2. The l-asparaginase activity in this study was 57% higher than the reference organism B. altitudinis BITHSP010. The l-asparaginase producing bacterium B. altitudinis KB1 from a marine source in Kerala can produce asparaginase, which can be utilised for biotechnology applications.

Whole genome sequence data of Bacillus australimaris strain B28A, isolated from Marine Water in India

Bacillus genus members are dominant in the Eastern Arabian Sea and are known for producing many industrial enzymes. Bacillus australimaris B28A, isolated from seawater, had an enzymatic activity. Here, the whole genome sequence of Bacillus australimaris B28A is reported. The 3,766,107-bp genome, with a GC content of 41.6%, comprised 3936 protein-coding genes, seven ribosomal RNA, and 75 transfer RNA. Several bioactive secondary metabolite genes in the genome, including surfactin, lichenysin, bacillibactin, bacilysin, paenilamicin, fengycin, and carotenoid, were identified using antiSMASH. The 1396 proteins were predicted using RAST, including asparaginase enzyme: an anticancer enzyme. Sequence data have been deposited in the DDBJ/ENA/GenBank database under the accession number JAGQFH000000000. The version described in this paper is JAGQFH000000000.1. The BioProject ID in the GenBank database is PRJNA670955. The raw data is publicly available at "https://www.ncbi.nlm.nih.gov/sra/SRR14203888".

Evaluating the Performance of Active Queue Management Using Discrete-Time Analytical Model

Congestion in networks is considered a serious problem; in order to manage and control this phenomena in early stages before it occurs, a derivation of a new discrete-time queuing network analytical model based on dynamic random early drop (DRED) algorithm is derived to present analytical expressions to calculate three performance measures: average queue length (Qavg,j), packet-loss rate (Ploss,j), and packet dropping probability (pd(j)). Many scenarios can be implemented to analyze the effectiveness and flexibility of the model. We compare between the three queue nodes of the proposed model using the derived performance measures to identify which queue node provides better performance. Results show that queue node one provides highest Qavg,j, Ploss,j, and (pd(j)) than queue nodes two and three, since it has the highest priority than other nodes. All the above results of performance measure are obtained only based on the queuing network setting parameters.