PSO optimization on backpropagation for fish catch production prediction

Abstract

Global climate change is an issue that is enough to grab the attention of the world community. This is mainly because of the impact it has on human life. The impact that is felt also occurs in waters on the South Kalimantan region. This is of course can disrupt the productivity of fish in the marine waters of South Kalimantan. This study aims to make fish catch production prediction models based on climate change in the South Kalimantan Province because the amount of productivity of marine fish has fluctuated. This study uses climate data as input and fish production as output which is divided into two, namely training and testing data. Then the prediction is conducted using Backpropagation method combined with Particle Swarm Optimization method. The results of the study produced a prediction model with RMSE of 0.0909 with a combination of parameters used, namely, C1: 2, C2: 2, w: 0.7, learning rate: 0.5, Momentum: 0.1, Particles: 5, and epoch: 500. While the model used when predicting testing data produces RMSE of 0.1448.

Swarm Optimisation to Model the Surface Roughness of an AISI 4340 Turning using the Hot Machining Process

“…The swarm approach has been used in various research domains, including driverless cars, fisheries, underwater vehicles and optimal economic load dispatch [20][21][22][23][24][25]. The PSO technique is based on stochastic processes and is built around a population of organisms/particles.…”

“…The designed system remains unchanged under various machining conditions, and the user only needs to prepare the new knowledge base. To configure a PSO algorithm correctly, some extra parameters must be chosen (Table 2), which in our instance are as follows [23,27,28]: Given that PSO was used to model the machining process, PSO parameters were used to regulate the optimisation method. The findings were to be displayed in the form of coefficients C0, a, b and c, which defined the specific combination and weight factor of each independent machining parameter according to the design of the PSO method [20,21].…”

Medical remotely caring with COVID-19 virus infected people using optimized wireless arm tracing system

“…It is based on praline swarm intelligent and is inspired by the animals' behaviors as birds or fish. PSO is an easy type of optimization algorithms that is used in many applications in different fields as engineering and science for example, data mining, image processing, machine learning and various other fields [21,22]. Initially, PSO introduced by James Kennedy and Russell C. Eberhert in 1995 [23].…”

Implementation multiple linear regresion in neural network predict gold price

“…Based on the research that has been done in predicting gold prices using ANN, it can provide output with good accuracy, so that it can be taken into consideration as well as important indicators of the economic sectors of various companies [4]. From the results of other prediction processes, ANN can produce maximum results in making predictions with a fairly low percentage of error values [5], [6]. The same research states that ANN is faster and has a high degree of accuracy in making [7], [8].…”