a. salajegheh

Today, with the proliferation of complex networks and their large amounts of data, researchers have great concerns about the accurate community detection methods. The difficulty in analyzing these networks stems from their enormous size and the complex relationships among the members of the networks. It is difficult to analyze the deep relationships and mechanisms by just looking at the whole. Traditional methods have some problems and limitations when analyzing these networks such as feature extraction, high reliance on the initial phase settings, computational complexity, neglect of network relationships and content. From the perspective of relationships and interactions between individuals, the environment of complex networks can be compared to a game in which nodes acting as players or agents may join or leave a community based on similar structural or semantic characteristics. Consequently, there is a strong tendency to use cooperative and non-cooperative games to detect communities. Moreover, the amalgamation of deep learning techniques and game theory has recently been proven to be highly effective in extracting communities. Deep learning techniques have demonstrated enhanced capability in feature engineering and automate the process. In this study, the authors make effort to detect rational and accurate communities based on structural and content features with the help of traditional approaches, deep learning, as well as cooperative and non-cooperative games. The efficiency of this study is demonstrated by experimental findings on real datasets, and confirming that it is able enough to identify those communities that are more meaningful.

Recently, network representation has attracted many research works mostly concentrating on representing of nodes in a dense low-dimensional vector. There exist some network embedding methods focusing only on the node structure and some others considering the content information within the nodes. In this paper, we propose HDNR; a hybrid deep network representation model, which uses a triplet deep neural network architecture that considers both the node structure and content information for network representation. In addition, the author's writing style is also considered as a significant feature in the node content information. Inspired by the application of deep learning in natural language processing, our model utilizes a deep random walk method to exploit inter-node structures and two deep sequence prediction methods to extract nodes' content information. The embedding vectors generated in this manner were shown to have the ability of boosting each other for learning optimal node representation, detecting more informative features and ultimately a better community detection. The experimental results confirm the effectiveness of this model for network representation compared to other baseline methods.

Forest fires can have a significant impact on water quality, especially in basins that provide drinking water. To investigate the effects of low, medium and high intensities of prescribed fire on runoff quality, 12 runoff plots were established in Kheyrud educational and research forest, north of Iran. In the first rainfall event after fire, six and 12 months after the fire, runoff samples were collected and transferred to the laboratory. Qualitative characteristics of the runoff were pH, electrical conductivity (EC), total suspended solids (TSS), total soluble solids (TDS) and total hardness (TH). The results showed that all the quality characteristics of the runoff after the fire increased compared to the control plots, but the differences among them were not significant. Most of these characteristics were in the high fire intensity, but as time passed after the fire, they had had a declining trend. The time elapsed after the fire had a significant effect only on TSS and TH. TSS values after six months and pH values after one year at low and medium fire intensities reached the level of control plots. TDS values immediately in the first event of precipitation in control, low, medium and high intensities were 22, 46, 53 and 78 mg/l, respectively. After one year, the values were 25, 36, 41 and 44 mg, respectively. Fire causes changes in the quality characteristics of runoff, element loss and reduced soil fertility. In order to reduce soil degradation, forest soil management measures after fire are necessary.

In order to study the effect of fire on some physical, chemical and biological characteristics of soil, 36 sample plots with three low, medium and high fire intensities and control treatments were set up in the Namkhaneh district of University of Tehran Forest. Soil samples were then taken from 0-10 cm layer in three phases (one day, 6 and 12 months after the fire) and some soil properties were studied. Data analysis was done by repeated measures ANOVA in SPSS software. The results showed that fire intensity had no significant effect on soil physical properties including bulk density, moisture content and soil texture. The results of soil chemical properties analysis showed that fire intensity had a significant effect on electrical conductivity, organic carbon, total nitrogen, carbon to nitrogen, phosphorus, calcium, potassium and magnesium ratios but had no significant effect on pH. In terms of biological characteristics, fire intensity had no significant effect on carbon microbial biomass but on microbial respiration and nitrogen microbial biomass. Time after fire had significant effect on moisture content, microbial respiration, microbial carbon and nitrogen, available phosphorus and potassium but had no significant effect on other physical, chemical properties. Since fire causes long-term depletion of soil nutrients, some post-fire preventive and rehabilitative measures in areas with moderate and high fire severities are required to restore soil functions more quickly.

Iran is among the 10 most vulnerable countries in the world. Among natural disasters which caused this vulnerability, flood has high priority. Paying attention to the vulnerability of the country as well as the vulnerability of cities against the natural disasters threats is one of the most important aspects of development planning. Hence, making decision in urban areas is one of the most important issues in modern management. The purpose of current research is to prioritize urban vulnerability to floods using Delphi method. The statistical population includes specialists who had sufficient knowledge and experience in the field of vulnerability management in urban areas. To carry out this research, initially the factors influencing urban vulnerability to floods was determined through literature review. After that effective indicators were ranked and selected using expert’s opinion and Delphi method. Based on the results, the main channel transmission capacity and flood control operations with 30.2%, drainage density with 29%, topographic conditions and land use of urban area with 28.6% were the most important criteria. The results of this study demonstrated that Delfi method is an appropriate and acceptable tool to development effective indicators in depicting of the cities vulnerability against floods.

Physical development of cities and expansion of its dimensions is one of the important forms of land use change that has many environmental, economic and social consequences. Changes in the natural pattern of drainage systems in urban and suburban areas, unprincipled use, development Impermeable surfaces are a major cause of urban flooding. The purpose of this study is to estimate the changes in discharge and runoff volume leading to the city as a result of increasing impermeable levels and urban development. In this regard, land use changes related to Abshouran watershed in three time periods have been tracked using Landsat images. EPA SWMM model was used to investigate changes in runoff peak flow. After calibrating the SWMM model based on the observed rainfall-runoff events, the calibration and validation results confirmed the conformity of the model simulations. Fitting of land use changes and corresponding floods during the study period shows that the area of urban areas in 2018 and 2001 compared to 1991 has increased by 189.68 and 60.79 percent, respectively. With increasing the percentage of urban impermeable surfaces, peak discharge and runoff volume for rainfall with a 20 year return period in 2018 compared to 1991, 198.82 and 188.04, respectively, and compared to 2001, 72.34 and 94.85 percent, respectively. Increased.

Spermatogenesis system is one of the most radiosensitive organs in the body. A usual therapeutic dose of radiation such as the conventional 2 Gy in each fraction of radiotherapy and lower doses seen in diagnostic radiology or a radiation disaster affect the process of spermatogenesis potently. Selenium and zinc are two important elements playing key roles in the development of sperms and also have radioprotective effects.

In this study aims to evaluate the radioprotective effect of zinc and selenium against radiation-induced mice testis injury.

In this experimental study, 30 mice were divided equally into 6 groups, including control selenium treated, zinc treated, radiation, radiation + selenium, radiation + zinc. Treatments started from 2 days before irradiation with 2 Gy cobalt-60 gamma rays. After 37 days, all mice were killed for histopathological evaluations.

Results showed that exposure to radiation caused a potent effect on spermatogenesis system. Treatment with selenium reversed these radiation effects potently, while zinc had some limited protective effects. Zinc treatment itself caused a detrimental effect on epididymis and, in combination with radiation, it leads to more damage to seminiferous tubules.

In contrast to previous studies that proposed zinc to protect spermatogenesis against various toxic agents, results of this study showed that although zinc may protect from some parameters, it potentiates radiation damage on seminiferous tubules and has a detrimental effect on the epididymis. By contrast, zinc and selenium could alleviate radiation-induced toxicity on the most of the evaluated parameters.

Land-surface models are widely used to simulate and predict the movement of momentum, energy, and matter between the biosphere and the atmosphere at local, regional and global scales. In this study, in order to evaluate the performance of this model in daily simulation of discharge in Gharesou watershed, using a spatial resolution of 0.125 degrees, satellite imagery and global databases to provide inputs for the model during the period from 01/01/2003 to 30/09/2008 for calibration and 01/01/1993 to 31/12/2003 for Validation. The results of the evaluation of the model's efficiency with the statistical indices of the Coefficient of Determination and Nash- Sutcliff efficiency coefficients were 0.75 and 0.73 in the calibration period, and 0.72 and 0.74, in the Validation period, indicate the high performance of the model. Generally, discharge simulated by the model, the trend of simulated and observed discharge is the same and due to the scales of VIC-3L and using of its in-watershed scales, it can be stated VIC-3L has a high performance.

Gully erosion is one of the most complicated and destructive kinds of watererosion. In order to prevent this erosion, the important factors of gully headadvancement must be recognized. For this case, this research has been done in one ofthe sub basins of Hable Rood drainage basin with name’s of Dehnamak in arid andsemiarid weather. Then, effective factors of gully head advancement of four models:Thompson, SCS(І), SCS(ІІ) and FAO, that are: gully head cut watershed area, claypercentage amounts, gully bottom slope percentage and other factors such as solublesalts percentage, silt percentage amounts and sand percentage amounts have beenminded. Meaning amount of each effective factor with gully head advancementamounts in stage of 95% in period of time 1956 to 2005 studied.Result of statistical studies shows, factors such as gully head cut watershed areawith correlation coefficient of 0.98, soluble salts percentage with correlationcoefficient of 0.97 and clay percentage amounts with correlation coefficient of 0.96,intermittently have a most meaning in stage of 0.95. After this, model with usingeffective factors, above mentioned, and SPSS software and two methods, Enter andBackward formed. Then, potential of gully head advancement in following years in theway of comparison estimated amounts by models and measure amounts by aerialphotos studied. Finally with regards to minute difference of 0.001% between twomethods Enter and Backward, Backward method proposed.