Sapientia
Repositório Científico da UAlg
Entradas recentes
Automatic ceramic identification using machine learning. Lusitanian amphorae and Faience. Two Portuguese case studies
Publication . Santos, Joel; Nunes, Diogo A.P.; Padnevych, Ruslan; Quaresma, José Carlos; Lopes, Martim; Gil, Joana; BERNARDES, João Pedro; Casimiro, Tania Manuel
This article presents a novel approach to classifying archaeological artefacts using machine learning, specifically deep learning, rather than relying on traditional, time-consuming human-based methods. By employing Convolutional Neural Networks (CNNs), this approach aims to expedite and enhance the identification process, making it more accessible to a wider audience. The study focuses on two types of artefacts- Roman Lusitanian amphorae (2nd-5th centuries) and Portuguese faience (16th-18th centuries)- chosen for their diversity. While Lusitanian amphorae lack decoration, Portuguese faience poses challenges with subtle colour variations. The study demonstrates the potential of this approach to overcome these hurdles. The paper outlines the methodology, dataset creation, and model training, emphasizing the importance of extensive data and computational resources. The ultimate objective of this research is to develop a mobile application that utilizes image classification techniques to accurately classify ceramic sherds and bring about a significant transformation in archaeological classification.
Physiology of spore formation for Bacillus probiotic production
Publication . Pawar, Lokesh; Singh, Arya; Chouhan, Nayan; Amer, Hadeer A.; Likitha, K.; Prasanthmadduluri, Naga; Singh, Soibam Khogen; Soltani, Mehdi
Probiotic Immune response Bacillus sp Physiology of Bacillus
Bacillus species are highly regarded as probiotics in aquaculture due to their positive impact on host health and disease prevention. This chapter provides a comprehensive overview of the physiological aspects and optimization strategies involved in the production of Bacillus spp. probiotics for aquaculture that contributes to the development of efficient and sustainable aquaculture practices. It highlights the significance of Bacillus spp. as probiotic, their potential in enhancing aquaculture productivity, and the importance of understanding the physiological characteristics of Bacillus spp. to optimize their growth and spore production. The chapter discusses several key factors that influence Bacillus spp. spore production and growth. It explores the effects of carbon sources, lignocellulosic growth substrates, nitrogen sources, medium pH, agitation and aeration, and microelements on the physiology and productivity of Bacillus species. Additionally, the chapter emphasizes the importance of selecting the appropriate cultivation method for scaled-up Bacillus spp. probiotic production in aquaculture. Different cultivation methods, such as batch, continuous, and fed-batch cultures, are evaluated, taking into account their impact on growth, spore formation, and overall probiotic yield. It enables the cultivation of high-quality probiotics with enhanced benefits, including disease prevention, improved nutrient utilization, and growth promotion, thereby enhancing the health and productivity of aquaculture organisms.
Integrating co-creation into crowdfunding: a comprehensive model for internet users’ intention to participate in tourism projects
Publication . Laachach, Abderrahim; Sadighha, Jinous; Azza, Yahya
This study aims to develop and validate a comprehensive model that elucidates the factors influencing internet users’ intentions to participate in crowdfunding campaigns for tourism projects. This research utilized a quantitative methodology, collecting data from 385 participants through an online survey as well as through offline methods. The analysis was conducted using Partial Least Squares Structural Equation Modeling (PLSSEM), enabling the examination of the proposed hypotheses and validation of the comprehensive model that identifies the factors affecting internet users’ intentions to engage in tourism crowdfunding projects. The results reveal that trust, perceived usefulness, perceived risk, communication quality, engagement, and social influences significantly predict internet users’ intentions to participate in crowdfunding tourism projects. Moreover, trust plays a crucial mediating role, mitigating the impact of perceived risk on participation. Additionally, crowdfunding platform ease of use only indirectly enhances users’ participation intention through trust. These findings underscore the importance of sharing transparent information about project’s potential risk and giving users access to secure crowdfunding processes to build trust towards the promotor and enhance users’ participation in tourism crowdfunding campaigns. This study advances the understanding of crowdfunding participation in the tourism sector by introducing a novel framework that integrates co-creation theory and social proof theory. The empirical validation of this comprehensive model offers valuable insights for practitioners and policymakers seeking to enhance the effectiveness of tourism crowdfunding campaigns.
Leveraging NLP and machine learning for English (L1) writing assessment in developmental education
Publication . Da Corte, Miguel; Baptista, Jorge
This study investigates using machine learning and linguistic features to predict placements in Developmental Education (DevEd) courses based on English (L1) writing proficiency. Placement in these courses is often performed using systems like ACCUPLACER, which automatically assesses and scores standardized writing assignments in entrance exams. Literature on ACCUPLACER’s assessment methods and the features accounted for in the scoring process is scarce. To identify the linguistic features important for placement decisions, 100 essays were randomly selected and analyzed from a pool of essays written by 290 native speakers. A total of 457 Linguistic attributes were extracted using COH-METRIX (106), the Common Text Analysis Platform (CTAP) (330), plus 21 DevEd-specific features produced by the manual annotation of the corpus. Using the ORANGE Text Mining toolkit, several supervised Machine-learning (ML) experiments with two classification scenarios (full and split sample essays) were conducted to determine the best linguistic features and bestperforming ML algorithm. Results revealed that the Naive Bayes, with a selection of the 30 highest-ranking features (21 CTAP, 7 COH-METRIX, 2 DevEd-specific) based on the Information Gain scoring method, achieved a classification accuracy (CA) of 77.3%, improving to 81.8% with 60 features. This approach surpassed the baseline accuracy of 72.7% for the full essay scenario, demonstrating enhanced placement accuracy and providing new insights into students’ linguistic skills in DevEd.
Relationship between landscape pattern and human disturbance in Serbia from 2000 to 2018
Publication . Quinta-Nova, Luís; Gómez, José Manuel Naranjo; Vulevic, Ana; Castanho, Rui Alexandre; Loures, Luis
This study intends to verify how the alteration of the landscape configuration, represented by different metrics of configuration and diversity, is related to the intensity of human disturbance. The objectives of the study are: (1) to quantify the change in land use/land cover (LULC) patterns and the degree of human disturbance in Serbia between 2000 and 2018, and (2) to study the relationship between LULC configuration and the impact resulting from human disturbance under different levels of intensity, to understand how changing trends in landscape pattern can serve as indicators to estimate landscape changes resulting from human actions. The Hemeroby Index (HI) was calculated to quantify the impacts on ecosystems resulting from disturbance caused by human actions. Based on the analysis of the variation in the value corresponding to the HI for the period between 2000 and 2018, the level of naturalness increased by only 5% of the territory of Serbia, with this change being verified mainly in SE Serbia. The landscape pattern was quantified using a set of LULC metrics. We used the Spearman method to identify the existing statistical correlations between the geometric parameters of the landscape and the HIs values. At the landscape level, the Mean Shape Index, Edge Density, Mean Patch Fractal Dimension, and Shannon Diversity Index show a strong negative correlation with HI. This correlation suggests that landscapes with greater structural complexity are good indicators of low levels of hemeroby. At the class level, Edge Density and Mean Patch Size correlate significantly with the HI for artificial surfaces, agricultural areas, forests, and semi-natural areas.