Using object-based image analysis (OBIA) techniques for land use-land cover classification (LULC) has become an area of interest due to the availability of high-resolution data and segmentation methods. Multi-resolution segmentation in particular, statistically seen as the most used algorithm, is able to produce non-identical segmentations depending on the required parameters. The total effect of segmentation parameters on the classification accuracy of high-resolution imagery is still an open question, though some studies were implemented to define the optimum segmentation parameters. However, recent studies have not properly considered the parameters and their consequences on LULC accuracy. The main objective of this study is to assess OBIA segmentation and classification accuracy according to the segmentation parameters using different overlap ratios during image object sampling for a predetermined scale. With this aim, we analyzed and compared (a) high-resolution color-infrared aerial images of a newly-developed urban area including different land use types; (b) combinations of multi-resolution segmentation with different shape, color, compactness, bands, and band-weights; and (c) accuracies of classifications based on varied segmentations. The results of various parameters in the study showed an explicit correlation between segmentation accuracies and classification accuracies. The effect of changes in segmentation parameters using different sample selection methods for five main LULC types was studied. Specifically, moderate shape and compactness values provided more consistency than lower and higher values; also, band weighting demonstrated substantial results due to the chosen bands. Differences in the variable importance of the classifications and changes in LULC maps were also explained.

The use of machine learning in real estate is quite new. When the working area is large, the factors affecting the price may vary according to the geographical regions and socioeconomic factors. It is thought that the price prediction performance of a model that will reflect these differences will be more successful than a general model. Unsupervised learning methods can be used both to increase performance and to show the variation of different factors affecting the price according to regions. With this aim, a hybrid model of X-Means clustering and CART decision trees was established in this study. This model successfully learned the geographical and physical variables that affect the price. The prediction performance of the model was compared with the direct capitalization method, which is the gold standard in the domain. The hybrid model has a superior performance over direct capitalization in terms of mean square error, root mean square error and adjusted R-Squared metrics. The scores were 72.86, 0.0057 and 0.978, respectively. The effect of clustering was also examined. Clustering increased the prediction performance by 36%.

The machine learning approaches are used in different domains for price prediction. Real estate price prediction comes to fore in recent years. However, most of the studies focus on the prediction performance and the factors affecting the price are often ignored. In this study, a C4.5 – CART model to predict the residential real estate prices is developed. This model is capable of predicting both numeric and categorical price for real estate properties. In addition, the factors affecting the price are reveled and analyzed in detail. The performance of the developed model is compared to Direct Capitalization model, which is used as a gold standard in the domain. Both models are tested on a dataset that includes updated real time data that is gathered by a web scraper. For numeric prediction, RMSE of the developed model is 13.169 and 358.69 for the Direct Capitalization model. KAPPA and accuracy is used for the categorical prediction. The model has 81% KAPPA and 88% accuracy.

Aims: Rice is one of the most applied irrigation waters applied plant among cultivated plants since it is grown in water or in saturated soil conditions in a significant portion of the growing season. This study was aimed to determine the effects of rice grown on water use, development, yield and some yield parameters using of both separately and jointly water retention barriers and subsurface drip irrigation methods in Enez, Edirne, Turkey in 2017.Methods and Results: Four different treatments were applied: ponding ((control(C)), subsurface drip irrigation system (SDI), ponding + water retention barrier (C+WRB) and SDI+WRB in this study. Subsurface drip irrigation laterals were placed 10 cm deep from the soil surface and the WRB was placed at 30 cm deep. Plants were irrigated with a constant water height of 10-15 cm on the soil in C and C+WRB treatments, 20% ± 5% of the available water retention capacity is consumed without exposure to water stress in SDI and SDI+WRB treatments, the field capacity is completed. WRB and SDI applications significantly affected the irrigation water amount, grain yield and yield components of rice. The yield and irrigation water amount according to the treatments varied between 321-715 kg da-1 and 751-2444 mm, respectively. Conclusions: In comparison to the control treatment, water saving was achieved by 27%, 50%, 69% in C+WRB, SDI and SDI + WRB treatments, respectively. Despite this, when marketable yield values are analysed according to the control treatment, it has increased by 10.8% in C+WRB, and decreased by 48.8% and 40.7% in SDI and SDI + WRB treatments. The water retention barriers and subsurface drip irrigation practices used in the study have been shown to reduce the use of irrigation water in rice production.Significance and Impact of the Study: It can be said that these methods have the potential to be used especially in arid and semi-arid regions where water is limited or inadequate.

Bu çalışmada; domates (Lycopersicum esculentum L. cv Full F1) bitkisinde, bitki su stresi indeksi (CWSI) ve Normalize Edilmiş Vejetatif Değişim İndeksi (NDVI) sensöründen elde edilen veriler kullanılarak su stresi düzeyinin, ayrıca CWSI ve NDVI değerleri ile bitkinin bazı fizyolojik özellikleri (stoma iletkenliği, yaprak su potansiyeli, yaprak oransal su içeriği ve klorofil) arasındaki ilişkilerin belirlenmesi amaçlanmıştır. Çanakkale ilinde 2017 yılında yürütülen çalışmada dört farklı sulama konusu (%100, %75, %50 ve %25) ele alınmıştır. Çalışma sonucunda, uzaktan algılama indekslerinin her ikisi de su stresi karşısında belirgin tepkiler vermiştir. Bu durumda her iki indeks de kullanılarak domatesin su stresinin başarılı bir şekilde belirlenebileceği söylenebilir. Buna ilaveten ölçümü zor, zaman alıcı ve bitkiye zarar verebilen fizyolojik ölçümlerin CWSI ve NDVI değerlerinin her ikisini de kullanarak yüksek doğrulukla tahmin edilebileceği sonucuna varılmıştır.

The study aimed to statistically analyze forest fire perceptions and cognitive deficits of children through drawings. Results showed that children's perceptions on forest fire were under desirable levels. Also, perception levels were significantly impacted by gender and grade level, whereas impact of school type was not significant. Since children's imaginations are found to be limited and most of forest fires are known to arise from human-induced activities, applied approaches are suggested to enlarge their vision on nature and forest fire with the purpose of raising more responsible, aware, and environmentally-conscious children, as a contributor to prevention of future forest fires.