Showing 7 results for Clustering
P. Shekari, M. Baghernejad,
Volume 9, Issue 4 (1-2006)
Abstract
Chenges in the soil characteristics is rather continuously. A method that takes this continuity into account would present a realistic pattern of soil distribution either in taxonomic or geographical space. The fuzzy set theory provides such an approach. In this study, the robustness of fuzzy clustering in soil pattern recognition was evaluated in a subcatchment of western Iran. The clustering carried out on the basis of minimization of an objective function in assigning membership values to each pedon in each fuzzy class. Fuzziness exponent values from 1.15 to 1.5 were used. The following validation of the resulted clusters (classes), optimal number of classes in whole, morphological and particle-size subsets were determined 8, 4, and 5 respectively. Plots of membership values across the landscape indicated class overlap and considerable contiguity. Considering low differentiation of these young soils and the high similarity among their properties, the method indicated a high capacity in recognizing different soil types over the study area. Furthermore, there was relationships between the soil fuzzy classes and landform. Thus, the method is capable in continuous classification, which could be so important in construction of continuous soil maps at low aggregation levels, e. g., pedon.
M. Nourzadeh, S. M. Hashemy, M. J. Malakouti,
Volume 15, Issue 57 (10-2011)
Abstract
Electrical conductivity and acidity of soil are the most important chemical factors of soil for agriculture. The nature of soil is in such a way that its change has a continuous form. The method that can take into account this continuity will be able to show a better picture of change in soil characteristics. Objectives of this research are to investigate the relations between measured electrical conductivity and soil acidity of Qom plain, and clustering, compare the clustering methods, determine the optimum numbers of cluster, and to zone the clusters in the study area. Accordingly, two fuzzy clustering methods FCM and GK, were used for data mining and clustering of 465 measured data. For estimating the appropriateness and comparison of two methods, some criteria including Partition Coefficient, Classification Entropy, Partition Index, Separation Index and Xie and Beni's Index were used. Data mining results showed that the optimum number of clusters for FCM and GK method was 15 and 17, respectively. After investigating the results of clustering and based on the criteria of appropriateness, it was indicated that GK was the best clustering method. According to this method, 295 data from 465 measured samples had more than 40 percent of membership function. So, 9 clusters from 17 clusters had more than 20 members. Then salinity-alkalinity zoning based on GK method to show the clusters distribution better in the study area was prepared. This prepared fuzzy map explained that most of Northwest and west belonged to cluster 1 and eastern parts of study area include belonged to cluster 17. Based on this, salinity-alkalinity and the ensuing soil degradation in east of study area is more likely than the west of it.
M. Abdi Dehkordi, A. A. Dehghani, M. Meftah, M. Kahe, M. Hesam, N. Dehghani,
Volume 18, Issue 68 (9-2014)
Abstract
In many water resource projects such as dams, flood control, navigability, river aesthetics, environmental issues and the estimation of suspended load have great importance. The complexity of sediment behavior and mathematical and physical model inability in simulation of sedimentation processes have led to the development of new technologies such as fuzzy logic which has the ability to identify nonlinear relationship between input and output variables. In this study, the application of fuzzy clustering algorithm in estimating the annual amount of sediment was studied. So, the corresponding data of flow and sediment discharge of Valykben station in kasilian basin during 1349-1350 till
1353-1354 period was daily determined. The data was divided in two groups i. e. 75% as training data and 25% for test data. Then, the efficiency of model was obtained by using statistical parameters such as correlation coefficient, nash-satklyf coefficient, mean square error root and variance ratio. The result showed that the classification of data on the annual time scale and use of fuzzy clustering algorithm can estimate 0.49 values of the measured annually suspended sediment transport. Furthermore, on the same scale of classification, i.e. annual scale, this value was obtained 0.19. Thus, using fuzzy clustering algorithm can lead to higher accuracy and reliability than rating curve method, which is suggested for estimating suspended sediment transport.
Z. Dehghan, S. S. Eslamian, R. Modarres,
Volume 22, Issue 4 (3-2019)
Abstract
Regionalization is one of the useful tools for carrying out effective analyses in regions lacking data or with having only incomplete data. One of the regionalization methods widely used in the hydrological studies is the clustering approach. Moreover, another effective factor on clustering is the degree of importance and participation level for each of these attributes. In this study, it was tried to use a broad range of attributes to compare their performance in regionalization. Then, according to the importance and role of each attribute in regionalization, the appropriate weight for each of the attributes in each category was determined using the principal component analysis (PCA) method, and the effect of this weighting in forming the homogenous regions was investigated by the Ward's clustering method. In this regard, the maximum 24-hour rainfall data of 63 meteorological stations located in Urmia Lake Basin (ULB) was used in this study during a time period of 30 years (1979-2008). Furthermore, seven categories of attributes were defined in order to regionalize the rainfall. The results showed that by considering different attributes and combining them with each other, a different clustering is obtained in each category in terms of the number of clusters and stations. Among seven categories of attributes, it was found that the geographical and climatic-geographical categories of attributes showed a more appropriate clustering over the ULB. Additionally, the weighting of attributes could have more effect on improving homogeneity and forming the independent clusters in most cases in terms of the scattering of station and how to locate over the basin.
H. Ghamarnia, F. Sasani, B. Yargholi,
Volume 23, Issue 1 (6-2019)
Abstract
Exploring the homogenous regions for site specific management is important, especially in the areas under different anthropogenic activities. This was investigated using multi-way analysis including Factor Analysis, Hierarchical Clustering Analysis and k means in the areas under long-term wastewater irrigation over a period of more than 40 years, in Shahre Rey, south of Tehran. By using Factor Analysis model, eight factors as eight geochemical groups were extracted to explain approximately 60% of the total variance related to 37 soil physicochemical properties. The most important groups included the nutrient elements (OM, OC and N), micronutrients (Mn and B), soil water adsorption capacity (Clay, Silt, Sand and CEC), salinity and osmotic pressure (EC, OP and TDS) and sodification (SAR and Na). The maximum values of Cophenet and Silhouette coefficients were equal to 0.77 and 0.83, respectively, dictating the selection of the average linkage approach in Hierarchical Clustering Analysis and three clusters in the k-average method with 19, 24 and 34 mapping units. The Thiessen Polygons method in GIS was applied to separate the geochemical groups in the form of mapping units. This output, which was, in fact, the combination of multi-way models and its visual representation in GIS under separated mapping units of study area, could present suitable management activities for the areas under each cluster.
H. Ahmadzadeh, A. Fakheri Fard, M.a Ghorbani, M. Tajrishy,
Volume 25, Issue 3 (12-2021)
Abstract
In drought risk management, the regional analysis of drought is significant. In this paper, this important issue is investigated by presenting the new hydrological regional drought index (RDI). For this purpose, the Ajichai basin was selected as the study area. First, the time series of the streamflow drought index (SDI) was calculated for each of the hydrometric stations in the basin f regional analysis of hydrological drought. Then, to determine the homogeneous regions in terms of hydrological drought, the k-means method was used for clustering analysis. Based on the clustering results, 6 Homogeneous regions were identified in the basin. For each of these regions, the time series of the RDI index was calculated from 1365 to 1393. The results showed that during the study period in each of the regions 1, 2, 3, 4, 5, and 6, mild Wet and mild drought has occurred at 82.1, 80.1, 78.9, 83.3, and 84.3 percent of regions, respectively. Also, the total percentage of drought events (moderate and high) is higher than the total percentage of wet events (moderate and high) in all regions. So, during the study period, the total percentage of drought events (moderate and high) is more than twice the total percentage of wet events (moderate and high) in regions 2 and 3.
A. Shahbaee Kotenaee, H. Asakereh,
Volume 26, Issue 4 (3-2023)
Abstract
Precipitation is one of the most significant climatic parameters; its distribution and values in different areas is the result of complex linear and nonlinear relationships between atmospheric elements-climatic processes and the spatial structure of the earth's surface environment. Classification of data and placing them in small and homogeneous zones can be effective in improving the understanding of these complex relationships and their results. In the present study, zoning and analyzing the distribution of rainfall in Iran concerning environmental factors was performed using the annual precipitation data of 3423 synoptic, climatological, and gauge stations in the country during the period from 1961 to 2015 and the altitude, slope, aspect, and station density data. After standardization and preparation of the data matrix, the optimal number of clusters was determined and the data set was entered into the neural-fuzzy network model (ANFIS-FCM). The results showed that the values of R2 and MAE indices were 0.76 and 0.23, respectively which indicate the appropriate accuracy of the model. It was also found that in the four output zones of the model, environmental factors have a high impact on the spatial distribution of precipitation. In the first and third zones, the combination of high altitude and slope factors along with geographical proximity to precipitation systems has caused the average annual rainfall in these zones to be 318 and 181 mm, respectively. The mean annual rainfall has decreased to about 100 mm by the weakening of the role of environmental factors in the second and fourth clusters.
