Clustering divides the data available as bulk into meaningful, useful groups (Clusters) without any prior knowledge about the data. Cluster analysis provides an abstraction from individual data objects to the clusters in which those objects reside. It is a key technique in the data mining and has become an important issue in many fields. This paper presents a novel Fractional Lion Algorithm (FLA) as an optimization methodology for the clustering problems. The proposed algorithm utilizes the lion's unique characteristics such as pride, laggardness exploitation, territorial defence and territorial take over. The Lion algorithm is modified with the fractional theory to search the cluster centroids. The proposed fractional lion algorithm estimates the centroids with the systematic initialization itself. Proposed methodology is a robust one, since the parameters utilized are insensitive and not problem dependent. The performance of the proposed rapid centroid estimation is evaluated using the cluster accuracy, jaccard coefficient and rand coefficient. The quality of this approach is evaluated on the benchmarked iris and wine data sets. On comparing with the particle swarm clustering algorithm, experimental results shows that the clustering accuracy of about 75% is achieved by the proposed algorithm.

No abstract

Clustering find various application in the variety fields like, telecommunication, medical image processing, bioinformatics and so on. This high demand poses a challenge to the researchers to develop an effective and efficient clustering algorithm for grouping the data objects. Accordingly, literature presents various algorithms for data clustering using partitional-based approaches. This paper presents a new clustering algorithm, namely, MKF-Firefly which is developed by combining the multiple kernel-based objective function and firefly algorithm. The firefly algorithm finds the optimal cluster centroids using the multiple kernel-based objective function. The centroids obtained from the firefly algorithm are then utilized for clustering process. The proposed clustering process is evaluated using Rand coefficient, Jaccard coefficient and Clustering Accuracy on the two different datasets like iris and wine. The proposed MKF-firefly achieved the clustering accuracy of 97% on the iris dataset.

COVID’19 is an emerging disease and the precise epidemiological profile does not exist in the world. Hence, the COVID’19 outbreak is treated as a Public Health Emergency of the International Concern by the World Health Organization (WHO). Hence, an effective and optimal prediction of COVID’19 mechanism, named Jaya Spider Monkey Optimization-based Deep Convolutional long short-term classifier (JayaSMO-based Deep ConvLSTM) is proposed in this research to predict the rate of confirmed, death, and recovered cases from the time series data. The proposed COVID’19 prediction method uses the COVID’19 data, which is the trending domain of research at the current era of fighting the COVID’19 attacks thereby, to reduce the death toll. However, the proposed JayaSMO algorithm is designed by integrating the Spider Monkey Optimization (SMO) with the Jaya algorithm, respectively. The Deep ConvLSTM classifier facilitates to predict the COVID’19 from the time series data based on the fitness function. Besides, the technical indicators, such as Relative Strength Index (RSI), Rate of Change (ROCR), Exponential Moving Average (EMA), Williams %R, Double Exponential Moving Average (DEMA), and Stochastic %K, are extracted effectively for further processing. Thus, the resulted output of the proposed JayaSMO-based Deep ConvLSTM is employed for COVID’19 prediction. Moreover, the developed model obtained the better performance using the metrics, like Mean Square Error (MSE), and Root Mean Square Error (RMSE) by considering confirmed, death, and the recovered cases of COVID’19 for China and Oman. Thus, the proposed JayaSMO-based Deep ConvLSTM showed improved results with a minimal MSE of 1.791, and the minimal RMSE of 1.338 based on confirmed cases in Oman. In addition, the developed model achieved the death cases with the values of 1.609, and 1.268 for MSE and RMSE, whereas the MSE and the RMSE value of 1.945, and 1.394 is achieved by the developed model using recovered cases in China.

No abstract

Due to the rapid growth of the web, the process of collecting the relevant web pages based on the user query is one of the major challenging tasks in recent days. Hence, it is very complicated for the users to know the most relevant information even though various search engines are widely employed. To deal with users' trouble in identifying the relevant information from the web, we have proposed a meta-lion search engine to capture and analyze the ranking scores of various search engines and thereby, generate the re-ranked score results. Accordingly, LionRank, a lion algorithm-based meta-search engine is proposed for the re-ranking of the web pages. Here, different features like text based, factor based, rank based and classifier based features are used by the underlying search engines. In classifier based feature extraction, we have used the fuzzy integrated extended nearest neighbor (FENN) classifier to include the semantics in feature extraction. Moreover, an intelligent re-ranking process is proposed based on the lion algorithm to fuse the features scores optimally. Finally, the results of the proposed LionRank is analyzed with the web page database collected through four benchmark queries, and the quantitative performance are analyzed using precision, recall, and F-score. From the results, we proved that the proposed LionRank obtained the maximum F-score of 81% as compared with that of existing search engines like QuadRank, Outrank, Google, Yahoo, and Bing.

No abstract

No abstract