Evaluating Fishing Capacity Based on DEA and Regression Analysis of China’s Offshore Fishery

The analysis of offshore fishing capacity is of great significance and practical value to the sustainable utilization and conservation of marine fishery resources. Based on the 2004–2020 China Fishery Statistical Yearbook, data envelopment analysis (DEA) was applied for measuring fishing capacity using a number of fishing vessels, total power, total tonnage, and the number of professional fishermen as the input measures and the annual catch as the output measure. Capacity utilization had a calculated range from 80.7 to 100%, and its average is 93.5%. In the first four years of 2003–2007, the excess investment rate of fishing vessels, total tonnage, total power, and fishermen was low (<5%). There was a consistent sharp upward trend in 2007, a gradual downward trend from 2007 to 2015, and an upward trend after reaching a low point in 2015, with the highest gross tonnage of fishing vessels reaching 25.5%. Four regression models that incorporate machine learning algorithms are used, including Lasso, Ridge, KNN, and Polynomial Features. The goodness of fit for the four models was used as the evaluation index, and the offshore annual catch based on the evaluation index was proposed. The forecasting annual catch of the polynomial model can reach 0.98. Furthermore, a comparative simulation of the DEA incorporating the polynomial model was carried out. The results show that DEA can evaluate input factors under the conditions of a given range, and the polynomial model has more advantages in forecasting annual catches. Furthermore, the combined application of DEA and polynomial model was used to analyze and discuss the management policies of China’s offshore fishery, which can provide help and reference for future management.

Download Full-text

Prediction of Damage Accumulation Effect of Wood Structural Members under Long-Term Service: A Machine Learning Approach

Materials ◽

10.3390/ma12081243 ◽

2019 ◽

Vol 12 (8) ◽

pp. 1243

Author(s):

Zheng Li ◽

Duo Tao ◽

Mengwei Li ◽

Zhan Shu ◽

Songshi Jing ◽

...

Keyword(s):

Machine Learning ◽

Damage Accumulation ◽

Goodness Of Fit ◽

Machine Learning Algorithms ◽

Cycle Performance ◽

Model Parameters ◽

Timber Structures ◽

Structural Members ◽

Accumulation Effect

It is well known that wood structural members can stand a relatively heavy load in the short term but will gradually get weaker if the load is applied for a longer period. This phenomenon is caused by the damage accumulation effect in wood and should be appropriately considered during the design of timber structures. Although various formulation methods (also known as classical models) have been proposed to evaluate the damage accumulation effect in wood, the calibration of model parameters is very time-consuming. Our work proposes a novel method to deal with the damage accumulation effect in wood that involves the application of machine learning algorithms. The proposed algorithm considers a multi-objective optimization process with a combination of goodness-of-fit and complexity. Long-term experimental data of typical wood species are used for developing the machine learning based damage accumulation model. Compared with existing pre-formulated models, our model managed to reduce the complexity of the model structure and give sufficiently accurate and unbiased predictions. This study aims to provide a novel tool for evaluating the damage accumulation in wood structural members, and the proposed model can further support the life-cycle performance assessment of timber structures under long-term service scenarios.

Download Full-text

Novel GIS Based Machine Learning Algorithms for Shallow Landslide Susceptibility Mapping

Sensors ◽

10.3390/s18113777 ◽

2018 ◽

Vol 18 (11) ◽

pp. 3777 ◽

Cited By ~ 61

Author(s):

Ataollah Shirzadi ◽

Karim Soliamani ◽

Mahmood Habibnejhad ◽

Ataollah Kavian ◽

Kamran Chapi ◽

...

Keyword(s):

Machine Learning ◽

Sample Size ◽

Prediction Accuracy ◽

Goodness Of Fit ◽

Learning Algorithm ◽

Machine Learning Algorithms ◽

Landslide Susceptibility Mapping ◽

Sample Sizes ◽

Promising Alternative ◽

Statistical Measures

The main objective of this research was to introduce a novel machine learning algorithm of alternating decision tree (ADTree) based on the multiboost (MB), bagging (BA), rotation forest (RF) and random subspace (RS) ensemble algorithms under two scenarios of different sample sizes and raster resolutions for spatial prediction of shallow landslides around Bijar City, Kurdistan Province, Iran. The evaluation of modeling process was checked by some statistical measures and area under the receiver operating characteristic curve (AUROC). Results show that, for combination of sample sizes of 60%/40% and 70%/30% with a raster resolution of 10 m, the RS model, while, for 80%/20% and 90%/10% with a raster resolution of 20 m, the MB model obtained a high goodness-of-fit and prediction accuracy. The RS-ADTree and MB-ADTree ensemble models outperformed the ADTree model in two scenarios. Overall, MB-ADTree in sample size of 80%/20% with a resolution of 20 m (area under the curve (AUC) = 0.942) and sample size of 60%/40% with a resolution of 10 m (AUC = 0.845) had the highest and lowest prediction accuracy, respectively. The findings confirm that the newly proposed models are very promising alternative tools to assist planners and decision makers in the task of managing landslide prone areas.

Download Full-text

Health-Behaviors Associated With the Growing Risk of Adolescent Suicide Attempts: A Data-Driven Cross-Sectional Study

American Journal of Health Promotion ◽

10.1177/0890117120977378 ◽

2020 ◽

pp. 089011712097737

Author(s):

Zhiyuan Wei ◽

Sayanti Mukherjee

Keyword(s):

Health Behaviors ◽

Goodness Of Fit ◽

Adolescent Suicide ◽

Suicide Attempts ◽

Predictive Accuracy ◽

Warning Signs ◽

Machine Learning Algorithms ◽

Data Driven ◽

Cross Sectional ◽

Increased Risk

Purpose: Identify and examine the associations between health behaviors and increased risk of adolescent suicide attempts, while controlling for socio-economic and demographic differences. Design: A data-driven analysis using cross-sectional data. Setting: Communities in the state of Montana from 1999 to 2017. Selected Montana as it persistently ranks among the top 3 vulnerable states in the U.S. over the past years. Subjects: Selected 22,447 adolescents of whom 1,631 adolescents attempted suicide at least once. Measures: Overall 29 variables (predictors) accounting for psychological behaviors, illegal substances consumption, daily activities at schools and demographic backgrounds were considered. Analysis: A library of machine learning algorithms along with the traditionally-used logistic regression were used to model and predict suicide attempt risk. Model performances—goodness-of-fit and predictive accuracy—were measured using accuracy, precision, recall and F-score metrics. Additionally, χ2 analysis was used to evaluate the statistical significance of each variable. Results: The non-parametric Bayesian tree ensemble model outperformed all other models, with 80.0% accuracy in goodness-of-fit (F-score: 0.802) and 78.2% in predictive accuracy (F-score: 0.785). Key health-behaviors identified include: being sad/hopeless ( p < 0.0001), followed by safety concerns at school ( p < 0.0001), physical fighting ( p < 0.0001), inhalant usage ( p < 0.0001), illegal drugs consumption at school ( p < 0.0001), current cigarette usage ( p < 0.0001), and having first sex at an early age (below 15 years of age). Additionally, the minority groups (American Indian/Alaska Natives, Hispanics/Latinos) ( p < 0.0001), and females ( p < 0.0001) are also found to be highly vulnerable to attempting suicides. Conclusion: Significant contribution of this work is understanding the key health-behaviors and health disparities that lead to higher frequency of suicide attempts among adolescents, while accounting for the non-linearity and complex interactions among the outcome and the exposure variables. Findings provide insights on key health-behaviors that can be viewed as early warning signs/precursors of suicide attempts among adolescents.

Download Full-text

TINGKAT KAPASITAS PENANGKAPAN ARMADA PUKAT CINCIN PEKALONGAN BERDASARKAN PADA MUSIM PENANGKAPAN

Jurnal Penelitian Perikanan Indonesia ◽

10.15578/jppi.15.4.2009.313-320 ◽

2017 ◽

Vol 15 (4) ◽

pp. 313

Author(s):

Hufiadi Hufiadi ◽

Eko Sri Wiyono

Keyword(s):

Technical Efficiency ◽

Capacity Utilization ◽

Purse Seine ◽

Data Envelopment ◽

Monsoon Period ◽

Utilization Factor ◽

The North ◽

Fishing Capacity ◽

Java Sea ◽

Using Data

Pengelolaan kapasitas penangkapan berikut metode pengukurannya sudah menjadi isu penting dalam pengeloiaan sumber daya pelagis kecil di Laut Jawa. Alternaiif pendekatan melalur pengendalian faktor faklo nputyang tidak efisien yang digunakan dalam usaha penangkapan. Tujuan penelitian ini adalah mengukur tingkat efisiensi teknis dan pemanfaatan kapasitas alat tangkap pukat cincin di Pekaiongan. Efisjensi penangkapan dan pemanfaatan kapasitas dari alat tangkap pukai cincin yang dikaji dianalisis berdasarkan pada musim penangkapan. Hasil penelitian menunjukkan bahwa pemanfaaian kapasitas jauh melebihr dari nilar optimumnya aiau telah terjadi kelebihan kapasitas pemanfaatannya berlangsung terulama pada musim barat. Pemanfaatan kapasitas penangkapan berada pada tingkat yang optimum hanya terjadi pada musinr peralihan I (musim barat ke musim timur). The management of fishing capacity along with measurement methods becomes an inlporlant issue in the small pelagic fisheies management in Java Sea, through altemative approach to contrcl inefficient input factors used in fishing activities. The objectives of tltis study are Io measure lhe level of technical efficiency and utilization capacity of purse seine in Pekalongan. The technical efficiency and utilization capacity were analyzed based on seasons. The fishing efficiency tneasurement was pertormed by using data envelopment analysis. The results show that purse seine ln Pekalongan on r/i/est season mean that fishing capacity utilization factor of the Pekalongan purse serne during the north wesf monsocn peiod exceed the optimum value or excess fishing capacity has been occureed during thai season. Frshing caDacity utilizaton for optmum levet only on n the rnter monsoon period was between March to May.

Download Full-text

Fishing Effort and Associated Catch per Unit Effort for Small-Scale Fisheries in the Mozambique Channel Region: 1950–2016

Frontiers in Marine Science ◽

10.3389/fmars.2021.707999 ◽

2021 ◽

Vol 8 ◽

Author(s):

Dirk Zeller ◽

Gabriel M. S. Vianna ◽

Matthew Ansell ◽

Angie Coulter ◽

Brittany Derrick ◽

...

Keyword(s):

Fishing Effort ◽

Small Scale ◽

Catch Per Unit Effort ◽

Important Species ◽

Small Scale Fisheries ◽

Mozambique Channel ◽

Fishing Vessels ◽

Unit Effort ◽

Fishing Capacity ◽

Channel Region

The Mozambique Channel region in East Africa has diverse marine ecosystems and serves as a migratory corridor for economically important species. Local and foreign industrial fisheries operate in the Mozambique Channel, but regional small-scale fisheries are the crucially important fisheries that provide food security, livelihoods, and economic opportunities for rural coastal communities. This study reconstructed and investigated trends in the fishing effort and catch per unit effort (CPUE) of small-scale marine fisheries in four Exclusive Economic Zones (EEZ) that constitute the Mozambique Channel, i.e., Union of Comoros, Madagascar, Mayotte, and Mozambique, from 1950 to 2016. Effective fishing effort for small-scale fisheries in the form of fishing capacity in kWdays (i.e., kilowatt days) was derived using the number, length, motorization (engine power) by fishing vessels, as well as an approximate human-powered equivalent for shore-based fishers without vessels, as well as days of fishing per year. Effective small-scale fishing effort in the Mozambique Channel increased by nearly 60 times from just over 386,000 kWdays in 1950 to over 23 million kWdays in 2016. Correspondingly, the overall small-scale CPUE, based on previously and independently reconstructed catch data declined by 91% in the region as a whole, from just under 175 kg⋅kWday–1 in the early 1950s to just over 15 kg⋅kWday–1 in recent years. All four EEZs showed the strongest declines in the small-scale CPUE in the earlier decades, driven by motorization and growth in vessel numbers impacting effective fishing effort. Increased motorization combined with a substantial growth in overall vessel numbers were the drivers of the increasing fishing effort and decreasing CPUE, and clearly suggest that continuing to increase the fishing capacity of small-scale fisheries in the absence of effective and restrictive management actions may exacerbate overexploitation risk.

Download Full-text

Managing fishing power: the case of Alaska red king crab (Paralithodes camtschaticus)

Canadian Journal of Fisheries and Aquatic Sciences ◽

10.1139/f03-138 ◽

2004 ◽

Vol 61 (1) ◽

pp. 43-53 ◽

Cited By ~ 5

Author(s):

Genevieve Briand ◽

Thomas Heckelei ◽

Scott C Matulich ◽

Ron C Mittelhammer

Keyword(s):

Simulation Model ◽

Capture Efficiency ◽

Red King Crab ◽

Paralithodes Camtschaticus ◽

Bristol Bay ◽

King Crab ◽

Vessel Size ◽

Fishing Vessels ◽

Fishing Capacity ◽

One Year

Because of concern about the inability to manage the Bristol Bay red king crab (Paralithodes camtschaticus) fishery in Alaska and, in particular, to use in-season fishery performance to close the fishery at or near the preseason guideline harvest level, increasingly stringent pot limits were adopted to elongate the collapsing seasons. This paper provides a rigorous examination of the effect that pot limits had on season elongation and whether a redistribution of wealth occurred between large and small fishing vessels as a result of the policy. A simulation model of the fishery shows that pot limits did not elongate the season sufficiently to improve in-season management. Moreover, the policy allowed vessels to capture efficiency gains arising from an industry-wide reduction in fishing capacity. Both vessel size classes benefited from mutual gear reduction in all years except 1992. Redistribution of wealth was found to occur only in one year of the five years examined.

Download Full-text

Trend and Predictive Analytics of Dengue Prevalence in Administrative Region

Handbook of Research on Applications and Implementations of Machine Learning Techniques - Advances in Computational Intelligence and Robotics ◽

10.4018/978-1-5225-9902-9.ch013 ◽

2020 ◽

pp. 236-262

Author(s):

Kannimuthu Subramanian ◽

Swathypriyadharsini P. ◽

Gunavathi C. ◽

Premalatha K.

Keyword(s):

Machine Learning ◽

Urban Areas ◽

Goodness Of Fit ◽

Predictive Analytics ◽

Learning Algorithms ◽

Public Health Problem ◽

Viral Disease ◽

Machine Learning Algorithms ◽

Severe Dengue ◽

Major Public Health Problem

Dengue is fast emerging pandemic-prone viral disease in many parts of the world. Dengue flourishes in urban areas, suburbs, and the countryside, but also affects more affluent neighborhoods in tropical and subtropical countries. Dengue is a mosquito-borne viral infection causing a severe flu-like illness and sometimes causing a potentially deadly complication called severe dengue. It is a major public health problem in India. Accurate and timely forecasts of dengue incidence in India are still lacking. In this chapter, the state-of-the-art machine learning algorithms are used to develop an accurate predictive model of dengue. Several machine learning algorithms are used as candidate models to predict dengue incidence. Performance and goodness of fit of the models were assessed, and it is found that the optimized SVR gives minimal RMSE 0.25. The classifiers are applied, and experiment results show that the extreme boost and random forest gives 93.65% accuracy.

Download Full-text

Application of Machine Learning to Mortality Modeling and Forecasting

Risks ◽

10.3390/risks7010026 ◽

2019 ◽

Vol 7 (1) ◽

pp. 26 ◽

Cited By ~ 8

Author(s):

Susanna Levantesi ◽

Virginia Pizzorusso

Keyword(s):

Machine Learning ◽

Stochastic Models ◽

Goodness Of Fit ◽

Mortality Rates ◽

Machine Learning Algorithms ◽

Stochastic Mortality ◽

Life Insurers ◽

Stochastic Mortality Models ◽

Mortality Models ◽

Out Of Sample

Estimation of future mortality rates still plays a central role among life insurers in pricing their products and managing longevity risk. In the literature on mortality modeling, a wide number of stochastic models have been proposed, most of them forecasting future mortality rates by extrapolating one or more latent factors. The abundance of proposed models shows that forecasting future mortality from historical trends is non-trivial. Following the idea proposed in Deprez et al. (2017), we use machine learning algorithms, able to catch patterns that are not commonly identifiable, to calibrate a parameter (the machine learning estimator), improving the goodness of fit of standard stochastic mortality models. The machine learning estimator is then forecasted according to the Lee-Carter framework, allowing one to obtain a higher forecasting quality of the standard stochastic models. Out-of sample forecasts are provided to verify the model accuracy.

Download Full-text

Short Term Prediction of PV Power Output Generation Using Hierarchical Probabilistic Model

Energies ◽

10.3390/en14102822 ◽

2021 ◽

Vol 14 (10) ◽

pp. 2822

Author(s):

Dongkyu Lee ◽

Jae-Weon Jeong ◽

Guebin Choi

Keyword(s):

Machine Learning ◽

Em Algorithm ◽

Power Output ◽

Goodness Of Fit ◽

Weather Forecasting ◽

Probabilistic Approach ◽

Machine Learning Algorithms ◽

Natural Energy ◽

The Em Algorithm ◽

Short Term Prediction

Photovoltaics are methods used to generate electricity by using solar cells, which convert natural energy from the sun. This generation makes use of unlimited natural energy. However, this generation is irregular because they depend on weather occurrences. For this reason, there is a need to improve their economic efficiency through accurate predictions and reducing their uncertainty. Most researches were conducted to predict photovoltaic generation with various machine learning and deep learning methods that have complicated structures and over-fitted performances. As improving the performance, this paper explores the probabilistic approach to improve the prediction of the photovoltaic rate of power output per hour. This research conducted a variable correlation analysis with output values and a specific EM algorithm (expectation and maximization) made from 6054 observations. A comparison was made between the performance of the EM algorithm with five different machine learning algorithms. The EM algorithm exhibited the best performance compared to other algorithms with an average of 0.75 accuracies. Notably, there is the benefit of performance, stability, the goodness of fit, lightness, and avoiding overfitting issues using the EM algorithm. According to the results, the EM algorithm improves photovoltaic power output prediction with simple weather forecasting services.

Download Full-text

KAPASITAS PENANGKAPAN PUKAT CINCIN MINI DI PEMALANG

Jurnal Penelitian Perikanan Indonesia ◽

10.15578/jppi.15.2.2009.171-178 ◽

2017 ◽

Vol 15 (2) ◽

pp. 171

Author(s):

Erfind Nurdin ◽

Tri Wahyu Budiarti

Keyword(s):

Data Envelopment Analysis ◽

Capacity Utilization ◽

Pelagic Fish ◽

Fish Stock ◽

Data Envelopment ◽

Small Pelagic Fish ◽

Fishing Capacity ◽

Java Sea ◽

Fishery Status ◽

Dea Method

Sumber daya ikan pelagis kecil di Laut Jawa merupakan komoditi perikanan yang penting dan bersifat strategis. Sejalan dengan perkembangan pemanfaatannya, tekanan penangkapan yang tinggi dialami oleh hampir seluruh kawasan, terlebih di wilayah tradisionil. Kapasitas penangkapan akhir-akhir ini menjadi isu penting dalam perikanan global, mengingat di berbagai kawasan terjadi kapasitas berlebih dan penurunan stok ikan. Salah satu cara untuk mengetahui status perikanan terkini dilakukan dengan mengukur kapasitas perikanan. Penelitian ini dilakukan pada bulan Mei dan Juli tahun 2007 dengan metode Data Envelopment Analysis. Nilai rata-rata pemanfaatan variabel input (VIU) untuk daya lampu dan bahan bakar menunjukkan nilai di atas 1 (optimum = 1) yang mengindikasikan adanya kelebihan penggunaan variabel sehingga dipandang perlu untuk melakukan pengurangan variabel input. Nilai pemanfaatan kapasitas (CU) lebih besar dari 0,5 hampir mencapai 60% dari 106 kapal contoh. Hal ini menunjukkan bahwa nilai CU kurang dari 1 mendominansi distribusi perolehan nilai CU yang dapat diartikan telah terjadi kelebihan kapasitas pemanfaatan untuk penangkapan pukat cincin mini dengan basis di Pemalang, sehingga perlu upaya pengurangan kapasitas. Small pelagic fish resource in Java sea is an important fishery commodity. Nowadays high fishing pressure is experienced by mostly fishing areas, particularly in traditional zone. Fishing capacity recently becomes important issue in global fishery, because over capacity and fish stock decreasing is occured in various areas. One method to know the fishery status is by measuring fishery capacity, which is categorized as over capacity, under capacity or optimal condition. This study was carried out on May and July 2007. The method applied in this measurement of fishing capacity was DEA method (Data Envelopment Analysis). The average of variable input utilization for light intensity and fuel showed value of more than 1 (optimum = 1), indicating the usage of excess variable that need to reduce the input variable. The value of higher than 0.5 of capacity utilization showed 60% from 106 ships sample. This indicates that the CU value of less than 1 dominated the distribution of capacity utilization value. For mini purse seine based in Pemalang showed over capacity. This means, it needs to reduce the effort of fishing capacity.

Download Full-text