Clustering of the self-organizing map reveals profiles of farm profitability and upscaling weights

Sulkava, Mika; Sepponen, Anne-Mari; Yli-Heikkilä, Maria; Latukka, Arto

doi:10.1016/j.neucom.2013.09.063

Cited by 11 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to R. Tadeusiewicz (Dudek-Dyduch et al 2009), a Kohonen network can detect relationships that would be overlooked if the traditional statistical grouping method was used. Kohonen neural networks are non-model sys- tems, they recognize the relationship between the studied individuals without any a priori assumptions as to their type, structure (Sulkava et al 2015). This approach is different from statistical surveys, in which it is necessary to initially formulate a hypothesis, determine the research sample and select the methods of their verification (Mohebi, Bagirov 2015).…”

Section: Results Of the Classification Of Methodsmentioning

confidence: 99%

Innovative Classification of Methods of the Future-Oriented Technology Analysis

Halicka

2016

Technological and Economic Development of Economy

View full text Add to dashboard Cite

Abstract. In the era characterized by significant dynamics of the environment traditional methods of anticipating the future, assuming the immutability of the factors affecting the forecasted phenomenon, may be in the long term ineffective. The modern approach of predicting the future of technology, taking into account the multidimensionality of the environment, is, among other things, the Future-Oriented Technology Analysis (FTA). Designing the FTA research procedure is a complex process, both in organizational and methodological terms. The catalogue of methods that can be used in this process is extensive and constantly open. However, in the source literature the rules for the selection of methods appropriate for the type of research were not specified. The ways of combining methods in the research process were also missing. The main aim of this article was to present the author's classification of methods of future-oriented technology analysis and indicate the possibilities of its application. In the text, using statistical methods and artificial neural networks, the classification of methods with the potential of exploitation in prospective technology analysis was carried out. Each of the received classes was analysed, the characteristics of particular groups of methods were selected, and authorial names characterizing the given classes were chosen. According to the author, the application of the proposed classification of methods of future-oriented technology analysis facilitates the design of the FTA research process. It will contribute to the systematization and standardization of the manner of selection of research methods. It will also allow for the selection of complementary methods.

show abstract

Section: Results Of the Classification Of Methodsmentioning

confidence: 99%

Innovative Classification of Methods of the Future-Oriented Technology Analysis

Halicka

2016

Technological and Economic Development of Economy

View full text Add to dashboard Cite

show abstract

“…SOM-based visualization and clustering have been reported in agriculture for disease detection, plant species detection [59] , biodiversity [60] , water resources utilization [42] and land use [61] among other applications. The application of SOM is limited in the social and economic context of agriculture [62] and to the best of our knowledge, this is the first attempt to use SOM in the research of farmers-led organizations and social capital assessment.…”

Section: Methodsmentioning

confidence: 99%

Data driven analysis of social capital in Farmer Producer Companies

Jayaraman,

Ramu,

Rajan

et al. 2023

Heliyon

View full text Add to dashboard Cite

“…Then, we examine the possible clustering of these factors into categories, based on clustering using a self-organising map (SOM). The SOM technique is particularly appropriate for clustering under conditions of a relatively small, non-linear (Allahyar et al , 2015; Kohonen, 2013; Sulkava et al , 2015) and random data set (Bação et al , 2004). The SOM technique offers improved performance in terms of accuracy and sensitivity when compared to other prevalent techniques such as k -means, hierarchical clustering and expectation maximising clustering (Abbas, 2008; Mangiameli et al , 1996; Mingoti and Lima, 2006).…”

Section: Introductionmentioning

confidence: 99%

Reverse logistics risk management: identification, clustering and risk mitigation strategies

Panjehfouladgaran

Lim

2020

View full text Add to dashboard Cite

PurposeReverse logistics (RL), an inseparable aspect of supply chain management, returns used products to recovery processes with the aim of reducing waste generation. Enterprises, however, seem reluctant to apply RL due to various types of risks which are perceived as posing an economic threat to businesses. This paper draws on a synthesis of supply chain and risk management literature to identify and cluster RL risk factors and to recommend risk mitigation strategies for reducing the negative impact of risks on RL implementation.Design/methodology/approachThe authors identify and cluster risk factors in RL by using risk management theory. Experts in RL and supply chain risk management validated the risk factors via a questionnaire. An unsupervised data mining method, self-organising map, is utilised to cluster RL risk factors into homogeneous categories.FindingsA total of 41 risk factors in the context of RL were identified and clustered into three different groups: strategic, tactical and operational. Risk mitigation strategies are recommended to mitigate the RL risk factors by drawing on supply chain risk management approaches.Originality/valueThis paper studies risks in RL and recommends risk management strategies to control and mitigate risk factors to implement RL successfully.

show abstract

Clustering of the self-organizing map reveals profiles of farm profitability and upscaling weights

Cited by 11 publications

References 18 publications

Innovative Classification of Methods of the Future-Oriented Technology Analysis

Innovative Classification of Methods of the Future-Oriented Technology Analysis

Data driven analysis of social capital in Farmer Producer Companies

Reverse logistics risk management: identification, clustering and risk mitigation strategies

Contact Info

Product

Resources

About