Wyniki wyszukiwania - Biblioteka Nauki

1

Links between Land Use Change, Land Surface Temperature and Partridge Distribution – An Analysis of Environmental Factors

100%

El Fallah K. , Ouhakki H. , Adiba A. , El Kharrim K. , Belghyti D.

|

tom Vol. 25, iss. 1

260--275

EN

The purpose of this research was to investigate the intricate connections among land use change, land surface temperature, and the distribution of partridges (Alectoris barbara), employing a comprehensive analysis of various environmental factors. Indeed, a variety of geospatial techniques have been used to analyze the spatio-temporal trends in temperature as a function of different classes of vegetation cover, and the geographic distribution of ecological niches for this species in Meknes province was modeled using Maxent 3.2 (Maximum Entropy) software. The study spanned a 22-year timeframe, from 2000 to 2021, during which alterations in each land use category were identified through the utilization of various sensors, incorporating Landsat 7 ETM+ and Landsat 8 OLI/TIRS in the analysis. The results induced a significant change in the land surface temperature (LST) with a range of 15.85–36.20°C, 12.76–38.24°C and 25.73–47.79°C for the years 2000, 2010 and 2020, respectively. However, this change was negatively correlated with the normalized difference vegetation index (NDVI). This decline in vegetation, in turn, manifests as a significant factor contributing to the diminution of partridge distribution. By empirically establishing these connections, the research not only underscores the impact of temperature-induced vegetation changes on partridge habitat but also enhances comprehension of the intricate ecological dynamics governing species distribution in the context of evolving land use patterns.

2

Named-entity recognition for Hindi language using context pattern-based maximum entropy

86%

Jain A. , Yadav D. , Arora A. , Tayal D. K.

|

2022

|

tom T. 23 (1)

81--115

EN

This paper describes a named-entity-recognition (NER) system for the Hindi language that uses two methodologies: an existing baseline maximum entropy-based named-entity (BL-MENE) model, and the proposed context pattern-based MENE (CP-MENE) framework. BL-MENE utilizes several baseline features for the NER task but suffers from inaccurate named-entity (NE) boundary detection, misclassification errors, and the partial recognition of NEs due to certain missing essentials. However, the CP-MENE-based NER task incorporates extensive features and patterns that are set to overcome these problems. In fact, CP-MENE’s features include right-boundary, left-boundary, part-of-speech, synonym, gazetteer and relative pronoun features. CP-MENE formulates a kind of recursive relationship for extracting highly ranked NE patterns that are generated through regular expressions via Python@ code. Since the web content of the Hindi language is arising nowadays (especially in health care applications), this work is conducted on the Hindi health data (HHD) corpus (which is readily available from the Kaggle dataset). Our experiments were conducted on four NE categories; namely, Person (PER), Disease (DIS), Consumable (CNS), and Symptom (SMP).

3

Fat Dormouse (Glis glis L.) Distribution Modeling in the Hyrcanian Relict Forests of Northern Iran

58%

Naderi M. , Kaboli M. , Ahmadi M. , Kryštufek B.

|

tom Vol. 64, nr 1

136--142

EN

The fat dormouse (Glis glis L.) is a small arboreal and extreme habitat specialist mammal that is tightly linked to the deciduous mixed forests dominated by Beech (Fagus orientalis) and oaks (Quercus sp.). Despite its status in Iran as a least concern species, dormice face high risk of extinction in some parts of Europe. The unique life history and large scale distribution of the species in the Palearctic region made it as an ideal model species. This habitat specialist rodent is particularly sensitive to size and connectivity of the forest patches. The fat dormouse shows very deep molecular and morphological divergence in its eastern most parts of its global distribution, in the Hyrcanian refugium of the Northern Iran. Therefore modeling its distributional range can leads to identify biodiversity hotspots and planning conservation activities. The meteorological data, land cover types, topographical variables and geo-referenced points representing geographical locations of the fat dormouse populations (latitude/longitude) in the study area were used as the primary MaxEnt model input data. The predictive accuracy of the Fat Dormouse ecological niche model was significant (training accuracy of 93.3%). This approach successfully identified the areas of the fat dormouse presence across the study area. The result suggests that the maximum entropy modeling approach can be implemented in the next step towards the development of new tools for monitoring the habitat fragmentation and identifying biodiversity hotspots.

4

Maxent Modelling for Distribution of Plant Species Habitats of Rangelands (Iran)

58%

Chahouki M. A. Z. , Sahragard H. P.

|

tom Vol. 64, nr 4

453--467

EN

Quantifying the pattern of habitat distribution for range plant species can assist sustainable planning of rangeland use and management. However, data of plant species distribution are often scarce and modeling of habitat distribution using commonly used models is difficult. In this study, the Maximum Entropy Method (MaxEnt) was used to model the distribution of plant habitat to find the effective variables in plant species occurrence in the Poshtkouh rangelands on Yazd province, central Iran. Maps of the environmental variables were generated using GIS and Geostatistics facilities. Accuracy of model output was assessed using area under the curve (AUC) of the receiver operating characteristic and keeping 30 percent of the data. Evaluation of model accuracy by AUC indicated good and acceptable predictive accuracy for all plant species habitats, except Artemisia sieberi which had high frequency. The predictive maps of Artemisia aucheri, Scariola orientalis — Astragalus albispinus, A. sieberi2 and A. sieberi — Zygophyllum eurypterum had fair agreement with their corresponding observed maps. In addition, the accuracy of S. orientalis — A. sieberi and Tamarix ramosissima predictive maps was low and the estimated conformity rate of prediction and observed maps was poor. In fact, due to differences in the optimal ecological range, level of agreement of predictive and observed maps at each site was different. MaxEnt was substantially excellent to predict distributions of plant species habitat with narrow ecological niches e.g. Rheum ribes — A. sieberi, Seidlitzia rosmarinus and Cornulaca monacantha. It can also perform well with fairly few samples due to employing regularization.