Location-Based Mobile Community Using Ants-Based Cluster Algorithm

Abstract

A location based service (LBS) is widely used on modern smartphone around the world asits built-in features. Each smartphone can access a google API or map. People can therefore share their location (latitude and longitude) among friends. Many LBS spots can easily form “location based mobile community (LBMC).” Since the nodes are mobile, the community group changes dynamically and is unstructured. Ant-based clustering algorithm is a special kind of optimization technique which is highly suitable for finding the adaptive clustering for volatile networks. This Paper Aims To form a location based mobile community (LBMC) by using Ant-based clustering algorithm. Due to the mobile type community, a vanishing community problem is also stated in this paper. Instead of redo a whole algorithm again, we modify an original algorithm by applying a pheromone concept to handle a change. Our algorithm is named as ABCA & VP which stands for Ant-Based Clustering Algorithm with Vanishing problem. More than 5,000 samples from their latitude and longitude coordinates in Thailand. From an experiment, K-means clustering work well in small data size and low number of clusters. In Small size of data between 50 and 1000, our algorithm runs battery when a number of clusters reach 15 clusters. In a big data size (between 1,000 and 5,000 samples), our algorithm outperforms K-means clustering when a number of clusters reach 20 clusters.

Indonesian pharmacy retailer segmentation using recency frequency monetary-location model and ant K-means algorithm

“…By combining ant clustering algorithm and K-means, the clustering problem can converge faster and more efficient. Also, previous study found that ant K-means is a stable performing algorithm for solving a various industrial clustering problems with more promising results [26], [27].…”

Instagram Ads Optimization Strategy to Increase Brand Awareness at CV Media Computindo

“…Furthermore, for the implementation of Instagram Ads at CV Media Computindo, it is important to adapt to the company's characteristics and goals [20,21]. By understanding the target audience, creating engaging content, and executing effective campaign management, CV Media Computindo can leverage the potential of Instagram Ads to enhance brand awareness and achieve desired outcomes.…”

Gold Prices Time-Series Forecasting: Comparison of Statistical Techniques

“…Simultaneously, the research employs traditional methods such as Simple Average (SA), Moving Average (MA), Simple Exponential Smoothing (SES), Double Exponential Smoothing (DES), and Triple Exponential Smoothing (TES) to compute average gold prices. These techniques provide a nuanced understanding of the broader spectrum of prediction methodologies, each contributing unique analytical perspectives [12][13][14] [15].…”

“…This research aims to determine the best algorithm for predicting gold prices using various methods such as Linear Regression (LR), Naive Bayes (NB), Simple Average (SA), Moving Average (MA), Simple Exponential Smoothing (SES), Double Exponential Smoothing (DES), and Triple Exponential Smoothing (TES). Linear regression is used to find a linear relationship between gold prices and other important variables [12][13], while naive bayes is used for classifying the class of an object based on its attributes [14] [15]. Other methods like SES, DES, and TES are employed to calculate the average gold prices from the collected data.…”