ARC 2023

EVALUATION OF PROTEIN AND LIPID OXIDATION OF OREOCHROMIS NILOTICUS HARVESTED FROM HADAPANAGALA RESERVOIR WHICH WERE STORED UNDER CHILLED TEMPERATURE (4-10 OC)

Thu, 12 Oct 2023 00:00:00 GMT

EVALUATION OF PROTEIN AND LIPID OXIDATION OF OREOCHROMIS NILOTICUS HARVESTED FROM HADAPANAGALA RESERVOIR WHICH WERE STORED UNDER CHILLED TEMPERATURE (4-10 OC) Yapa, Y.M.D.M; Nam, K.C; Abeyrathne, E.D.N.S The study investigated the effects of chilled storage (4-10°C) over a period of seven days on protein and lipid oxidation in Nile tilapia (Oreochromis niloticus) harvested from Hadapanagala reservoir. Fish samples (250 ± 50 g) were harvested and stored at 4-10°C for 07 days. Protein oxidation was assessed by analyzing carbonyl concentration (nmol/mg protein) and SDS- PAGE analysis while lipid oxidation was assessed by analyzing malonaldehyde concentration (mg MDA per kg of fish) and antioxidant capacity (%). Freshness indicators, hypoxanthine proportion and pH values were also measured. According to data, there were significant differences in carbonyl content, malonaldehyde concentration and antioxidant capacity with different time periods of storage temperature condition at 4-10°C (P<0.05). Carbonyl content which was the protein oxidation indicator was reached to its maximum from the first day reading (5.41 ± 0.26) nmol/ mg protein to (7.87 ± 0.11) nmol/ mg protein on the seventh day while antioxidant capacity of the samples was reached its minimum from the first day reading (68.82 ± 3.28 %) to (18.82 ± 5.14 %) on the seventh day indicating the oxidation. These findings lead to conclude that storage temperature time period have a significant effect on oxidation of proteins and lipids hence preserving fish meat at 4-10°C for longer time periods is not very effective to in the terms of preserving of nutrients and extending shelf life.

ERROR ANALYSIS ON THE WRITTEN-ENGLISH OF THE FIRST YEAR STUDENTS AT TRINCOMALEE CAMPUS

Thu, 12 Oct 2023 00:00:00 GMT

ERROR ANALYSIS ON THE WRITTEN-ENGLISH OF THE FIRST YEAR STUDENTS AT TRINCOMALEE CAMPUS Naveenraj, V.J This research paper involves in a study of error analysis on the written sentences in English produced by the first year students who prosecute B.A in Communication Studies at the Department of Languages and Communication studies in Trincomalee Campus. The research paper begins with a systematic review of the concepts and theories concerning Error Analysis which sets the platform for the theoretical base. Subsequently, the Error Analysis focuses on finding out the syntactical errors under the selected grammatical categories while searching for the possible reasons causing errors. Eventually, the researcher suggests that the language teaching lecturers should understand the process involved in learning a new language. Therefore, the lecturers should understand the root causes of errors while attempting to find treatment for errors. It is paramount that the lecturers must approach error-corrections in accordance with the students’ linguistic competence, their affective factors embedded with different ethnic background. This error analysis has been undertaken with an intention to bring awareness to the teaching staff to help students produce better writing with less grammatical or no grammatical mistakes.

MODEL ML: AN AUTOMATED PLATFORM FOR PRECISE AND EFFICIENT MACHINE LEARNING MODEL DEVELOPMENT

Thu, 12 Oct 2023 00:00:00 GMT

MODEL ML: AN AUTOMATED PLATFORM FOR PRECISE AND EFFICIENT MACHINE LEARNING MODEL DEVELOPMENT Diluka, T; Wijesinghe; Gayan Perera In this paper, we introduce the "Model ML" platform, an innovative and automated solution with the primary objective of democratizing the development of machine learning models. This platform is tailored to benefit organizations and individuals who may not have direct affiliations with the machine learning and AI domain. It achieves this objective through the integration of multiple algorithms and the provision of an intuitive user interface. Machine learning (ML) and artificial intelligence (AI) have gained widespread use in diverse fields such as finance, marketing, and data analysis. However, the high cost of computing resources and the scarcity of expertise in data cleaning, feature selection, model selection, and hyper parameter optimization have hindered the adoption of machine-learning applications by novices, students, and small organizations. Although AutoML (Automated Machine Learning) platforms have emerged to simplify the process of developing machine learning models, their technical complexity and lack of user-friendly interfaces present a significant challenge for most users. This paper elucidates the strategies to surmount the mentioned difficulties by designing an automated platform for developing machine learning models, namely Model ML, as a web and mobile application. The proposed platform encompasses several salient attributes: user-friendly interfaces, dataset encoding, tailored model training, statistical analysis, and data visualization for personalized input data, instant model testing using user interfaces, streamlined workflows, and transparency. In addition, it incorporates a leaderboard function that facilitates the assessment of the algorithms utilized during training, as well as the option to export trained models in various formats and access pre-trained machine-learning models commonly employed in the field. As a result, individuals lacking extensive programming skills can construct precise and efficacious models with greater adaptability.

SEA SURFACE TEMPERATURE FLUCTUATIONS AROUND THE SEYCHELLS CRW MAHE SITE ON SEYCHELLS ISLAND, THROUGHOUT A 25-YEAR PERIOD (1998–2022), CONSISTING OF CORAL REEF ECOSYSTEMS

Thu, 12 Oct 2023 00:00:00 GMT

SEA SURFACE TEMPERATURE FLUCTUATIONS AROUND THE SEYCHELLS CRW MAHE SITE ON SEYCHELLS ISLAND, THROUGHOUT A 25-YEAR PERIOD (1998–2022), CONSISTING OF CORAL REEF ECOSYSTEMS Silva, T.D.T; Thilakarathne, E.D.P.N The increase in sea surface temperature caused by global warming is one of the major threats to coral reefs. The Republic of Seychelles is one of the world's most isolated and aesthetically pleasing groups of islands, and it has coral beaches. Therefore, this study was designed to study the fluctuation of inter-annual and intra-annual sea surface temperature from 1998 to 2022 over the most prominent coral reef in the Seychelles, the CRW Mahe Site coral reef. The data was gathered using the Coral Health and Monitoring Program Portal online database, which provides daily data from the fixed sea surface temperature sensor of the relevant reef. There was an annual average sea surface temperature trend increasing by 0.02 °C per year. The maximum and minimum monthly average sea surface temperatures recorded in April and August were 29.74°C and 26.075°C, respectively. The maximum and minimum yearly average sea surface temperatures recorded in 2019 and 2000 were 28.53°C and 27.39°C, respectively. The maximum yearly positive and negative anomaly sea surface temperatures recorded in 2019 and 2000 were 0.54 °C and -0.59°C, respectively. Comparatively, high positive yearly anomalies were recorded in 1998 (0.26°C), 2007 (0.24°C), 2009 (0.23°C), 2015 (0.39°C), 2016 (0.27°C), 2019 (0.54°C), and 2020 (0.48°C) years. El Niño, La Niña, Indian Ocean Dipole, and global warming all affect the region's usual yearly pattern of sea surface temperature variations. As a result, coral reefs can face thermal stress, which could cause coral bleaching and decrease the ecosystem's health.