Browsing by Subject "maximum entropy"

Now showing 1 - 2 of 2

Access status: Open Access ,
Assessing Potential Distributions of Bird Endemic Species: Case Studies of Macrocephalon maleo and Rhyticeros cassidix and Their Threats
(Wydawnictwa AGH, 2024) Aldiansyah, Septianto; Risna; Saputra, Randi Adrian
Maleo and knobbed hornbill are bird species that are endemic on the island of Sulawesi, which is highly threatened by forest fires. Fires tend to destroy any affected species, however, it is not possible to survey the entire range of the original distribution of the two endemic bird species that are affected by forest fires due to practical constraints. Species distribution modeling using maximum entropy is considered to be an alternative to understanding the potential distribution area of species against the threat of forest fires. The prediction model from MaxEnt all have AUC values of greater than 0.70, which means that the model is good enough to classify the records of the presence of M. maleo and R. cassidix along with the past forest fires. The environmental variables that affect the distribution of M. maleo are its distance from hot water, rivers, and roads, while the distribution of R. cassidix is strongly influenced by its distance from roads, settlements, and rivers. Forest fire distribution is mostly influenced by soil type, land-use land cover, and rainfall. It is predicted that around 238,690 and 677,070 ha of the potential distribution of M. maleo and R. cassidix, respectively, are potentially disturbed and affected by forest fires. However, this number much greater outside conservation areas. The results of this study can be used by the government of the Republic of Indonesia (especially the Ministry of Environment and Forestry) for determining conservation actions for both species in the future.
Access status: Open Access ,
Named-entity recognition for Hindi language using context pattern-based maximum entropy
(Wydawnictwa AGH, 2022) Jain, Arti; Yadav, Divakar; Arora, Anuja; Tayal, Devendra K.
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).