Statistical Physical Properties of Four-Wave Mixing Optical Neural Network

Исследованы статфизические свойства оптической нейросети. Получены условия, при которых возможно обучение нейросети алгоритмом максимального правдоподобия. Исследование проведено на примере трехмерной модели Изинга, в которой последовательно добавляется дальнодействие так, что в пределе модель можно описывать теорией среднего поля. Получены аналитические оценки для критической температуры нейросети при учете взаимодействия со вторыми и третьими соседями. Данные оценки на всем интервале значений параметров взаимодействия хорошо согласуются с результатами, полученными методами Монте-Карло. Установлено, что с ростом числа положительных межсвязей величина критической температуры падает и алгоритм максимального правдоподобия может применяться практически без ограничений. The paper investigates the statistical physical properties of an optical neural network. The conditions for training a neural network by the maximum likelihood algorithm are identified. The study uses a three-dimensional Ising model, to which a long-range action is sequentially added so that in the limit the model can be described by the mean-field theory. Analytical estimates of the critical neural network temperature were obtained considering the interaction with the second and third-order neighbors. The estimates for the entire interval of the interaction parameters are in good agreement with the results obtained by Monte Carlo methods. It is found that as the number of positive interconnections increase, the critical temperature value decreases and the maximum likelihood algorithm can be applied virtually without any restrictions.

Land Cover Change and Coastal Sustainability in the Coastal Area of Kendari Bay, Southeast Sulawesi, Indonesia

Coastal areas provide invaluable resources which have important environment, economic and social value. These resources encourages growing population and development which induced rapid changes in coastal areas. This study aims to analyse the changes in land cover of the coastal areas of Kendari Bay to provide recent perspectives of how land cover has changed using Landsat TM and Landsat OLI images for the period of 1998, 2008 and 2018. The classified land cover classes are categorized as waterbodies, built-up, bareland, forest, wetland, vegetation and mangrove. The land cover map of each period was acquired from supervised classification using maximum likelihood algorithm in ArcGIS, then the land cover change was analysed through post-classification change detection of GIS-based method. . Accuracy assessment of classified images shows the overall accuracy is estimated as 88.71%, 85.81% and 91.61%, and overall Kappa coeffient statistical values of 0.87, 0.83 and 0.90 for the year 1998, 2008 and 2018 respectively. This study found that there was significant land cover change in the coastal areas of Kendari Bay. It was dominated by the expansion of built-up areas and bareland by 55% and 469.77% respectively, which was gained from the conversion of vegetation and wetland. Meanwhile, considerable reduction were shown in mangrove, wetland, forest and vegetation which have declined by 48.65%, 43.39%, 38.72% and 27.20%. Analysing land cover change is an effective way to understand the dynamics of land cover in coastal areas, and can be used for future land use planning and policies.

The impact of mangrove damage on tidal flooding in the subdistrict of Tugu, Semarang, Central Java

Expansion of industrial areas, aquaculture, settlements, and limited knowledge of the community about the function of mangroves allegedly led to the conversion of mangrove functions in the early 1990s. This study aimed to map the condition of mangroves from 1988, 1990, 1995, 2008, to 2021 and their effect on the widespread of tidal flooding in three villages (Mangkang Kulon, Mangunharjo, Mangkang Wetan) in Tugu subdistrict, Semarang City. The research method was carried out by using spatiotemporal analysis of Landsat 5 and Landsat 8 imagery through the supervised approach (Maximum Likelihood algorithm). In order to map the correlation of mangrove damage with the widespread impact of tidal flooding, an overlay analysis of land use maps was carried out in 1988, 1990, 1995, 2008 and 2021. The results of the study showed that mangrove damage is correlated with the widespread of tidal flooding that drowns settlements, ponds, and agricultural land. Data analysis showed that the mangrove area in three villages has decreased from 1988 to 2021, covering an area of 242.66 ha. This condition is one of the triggers for the increase in tidal flooding area from 1988 to 2021, covering an area of 253.135 ha. As a natural barrier to prevent abrasion and tidal flooding, mangrove conservation is very necessary, considering the impact of tidal flooding on the coast of Semarang City is increasingly widespread.

Assessing the Dynamics of Urban Growth and Land Use Changes in Jeli Using Geospatial Technique

Land use change pattern in Jeli is been given a focus as part of tool for land planning anddevelopment. An increasing of population in Jeli make this study is relevant to aid an understandingon land use changes in this area due to the demanded for development and rapid land utilisation.Land use change pattern can be obtained via geospatial technique by Geographical InformationSystem (GIS) together with satellite imagery analysis. In this study, land use maps produced fromsupervised classification using maximum likelihood algorithm give a high accuracy of 92.05%.From classified land use images, urban expansion index (SI) were then calculated from year 1984to 2012. Pearson’s correlation analysis discovered a strong negative correlation between forest andagricultural areas which reflected that with increased of agriculture area, forest reserve was founddepleted. The SI of Jeli showed a low to rapid growth in 1994 and 1997 to 2012 respectively. Thestudy reflected the progressive development in Jeli promotes sustainable usage of forest area wheneffort of reforestation was carried out forming a co-existence land use category between forest andagriculture.

Sentinel-2 imagery usage on environmental monitoring of land use and occupation in a microwatershed in Central Brazil

In the last decades, sustainability concerns have increased the demand for projects and strategic plans that integrate economic and social aspects, reducing environmental impacts. In this sense, this study aims to monitor land-use adequacy in the Ribeirão Extrema microwatershed, Distrito Federal, based on cross-mapping between land-use and occupation in 2019 and agricultural aptitude map through Geographic Information Systems and Remote sensing. To this end, a hypsometric and thematic database was prepared for the region. Besides, we acquired an image from the Sentinel-2 orbital sensor of October 2019. The image was subjected to classification regarding land-use and occupation, using the MAXVER (maximum likelihood) algorithm. It was observed that 80% of use in 2019 was related to agricultural activities. Kappa index validation reached 81% accuracy. Based on the methodology, we identified 62.33% of agricultural activities occur into its capacity; 4.33%, were used above capacity, causing environmental degradation, especially in permanent preservation areas. The application of the technique was considered satisfactory because the adequacy of land-use in the studied microwatershed could be assessed in order to pursue sustainable development. Continuous analyzes can improve results.

The compost metagenome as a source of T4 bacteriophage pyrimidine dimer glycosylase homologues

Compost is a promising source of thermotolerant enzymes for their application in biotechnology. Homologues of bacteriophage T4 DNA glycosylase can find their application in pharmaceuticals and perfumery. Five homologues of glycosylase of pyrimidine dimers of bacteriophage T4, a product of the denV gene, were found by comparing using the DELTA-BLAST algorithm with the compost metagenome proteins. Phylogenetic analysis of the found sequences of enzyme homologues was carried out using the Maximum Likelihood algorithm in the MegaX software package. Thus, an interesting spectrum of promising proteins, homologues of the repair enzyme, DNA glycosylase of pyrimidine dimers of bacteriophage T4, was found. After structural modeling, they can be tested for their thermal stability and tested as a basis for therapeutic and prophylactic drugs.