ASEAN Journal of Scientific and Technological Reports

A Review Of Adaptive Multi-Sensor Grids in IOT-Enabled Smart City Development

2026-04-10T10:10:41+07:00

Rapid urbanization and continuous technological growth have increased the demand for sustainable and intelligent city infrastructures. Smart cities use the Internet of Things (IoT) and adaptive multi-sensor grids to improve resource management and real-time monitoring. This study focuses on understanding how IoT-based adaptive sensor grids can improve the sustainability, resilience, and intelligence of future cities. It also highlights the major innovations, present applications, and challenges involved in implementing these systems. A systematic literature review was carried out using academic databases such as IEEE Xplore and Scopus. Research papers published between 2020 and 2025 were analyzed using keywords like “smart city,” “IoT,” and “multi-sensor grid.” The findings show that adaptive sensor grids form the core of smart city infrastructure. These grids continuously monitor traffic flow, energy consumption, air quality, and public health conditions. When combined with Artificial Intelligence (AI), they support real-time decision-making. This enables predictive energy management, dynamic traffic control, and efficient waste handling. Despite these benefits, several challenges remain. The major concerns include cybersecurity risks, data privacy issues, and lack of interoperability between systems. Overall, adaptive multi-sensor grids play a crucial role in shaping sustainable and human-centered smart cities. To achieve this vision, it is essential to develop open standards, secure data frameworks, and strong governance policies. These steps ensure that IoT-driven urban systems remain safe, resilient, and fair for all citizens.

Development of Pilgrimage Tourism as an Application for Supporting Local Religious People

2026-03-16T19:03:51+07:00

Hajj and Umrah are both Islamic pilgrimages in Mecca, Saudi Arabia, which are the main activities for Muslims. These are also required by Muslims in Thailand, and the trend is increasing every year. Typically, the number of pilgrims who can perform Hajj is limited by Saudi Arabia. In 2022, Thailand received a quota of 13,000 people, and the Committee for the Promotion of Hajj Affairs in Thailand will distribute the quota to entrepreneurs and religious leaders who have licenses to lead pilgrimage tourism. However, there are some difficulties in a pilgrimage tour, such as visa applications, accommodation bookings, flights, and so on. These processes need a lot of paperwork, so entrepreneurs and religious leaders will present to support pilgrims. There have been reports of pilgrimage tourism, including high service charges, poor management, and fraud by scammers. To the best of our knowledge, there is no digital platform to help local religious leaders organize the tour and provide online services for pilgrims. Therefore, we design and develop a novel application for pilgrimage tourism to facilitate religious leaders in handling paperwork for the Committee for the Promotion of Hajj Affairs in Thailand and support pilgrims in booking pilgrimage tours through a mobile web application, namely https://www.hajestic.com. The key development of the application is a user-centric, e-marketplace-based structure integrated to secure online payments and ensure compliance for users. In addition, we also designed 3 qualitative questionnaires based on the technology acceptance model (TAM) to guarantee the application meets users’ requirements and satisfaction.

Development of Rubber Farmer Groups Towards Good Agricultural Practices for Rubber in the form of Cup Lump Rubber in the Northern Region of Thailand

2026-05-03T21:45:26+07:00

This research aims to develop a group of rubber farmers in northern Thailand to adopt Good Agricultural Practices (GAP) for cup-lump rubber and to explore approaches to developing GAP for cup-lump rubber. The study focused on rubber farmers in Phrao District, Chiang Mai Province, using a participatory action research methodology. The research involved two steps. 1) assessing the farmers' capacity to manage rubber plantations according to Good Agricultural Practices (GAP) principles; and 2) proposing recommendations for developing Good Agricultural Practices (GAP) for cup-lump rubber farmers. The study found that in 2022, farmers had an average rubber plantation area of 3.31 Ha, generating an average income of 235,549.33 baht per year from rubber plantations, while incurring average plantation management expenses of 4,621.46 baht per year. Most farmers sold their products as dry cup lumps to their cooperative members, and most had a high potential for Good Agricultural Practices (GAP) in rubber (average 4.13). The development of rubber farmer groups in northern Thailand toward Good Agricultural Practices (GAP) for rubber in the form of cup lumps should involve promoting member participation, particularly in farmer group management, and establishing clear standards for rubber trading. This should be coupled with the development of knowledge and skills in rubber plantation management through training with the Rubber Authority of Thailand. Furthermore, product quality should be improved, participatory inspection systems implemented, fair marketing and profit-sharing plans promoted, and sustainability planning implemented. However, to successfully develop farmer potential, farmers must begin by shaping their attitudes toward the benefits of Good Agricultural Practices (GAP) and promoting knowledge and understanding of proper Good Agricultural Practices (GAP).

Spatiotemporal Assessment of Physicochemical Water Quality in the Al-Aziziyah–Al-Hafariya Water Treatment System, Wasit, Iraq

2026-05-09T21:38:21+07:00

This study aimed to evaluate the physicochemical characteristics of water within the Al-Aziziyah–Al-Hafariya unified water treatment and distribution system in Wasit Governorate, Iraq. Water samples were collected monthly from six locations representing the raw water source, treatment stages, and the distribution network from September 2025 to February 2026. Key parameters were analyzed, including temperature, turbidity, total suspended solids (TSS), total dissolved solids (TDS), electrical conductivity (EC), salinity, pH, dissolved oxygen (DO), biological oxygen demand (BOD), alkalinity, hardness, residual chlorine, sulfate, nitrate, and phosphate. The results showed significant temporal variations (P ≤ 0.05) in most parameters, while spatial variations were generally insignificant across all sampling sites. Temperature, TDS, EC, salinity, and hardness were higher during autumn and decreased in winter. In contrast, dissolved oxygen and nutrient concentrations increased during colder months. Turbidity and TSS exhibited seasonal fluctuations. Critically, residual chlorine fell to zero at all six sampling sites during January and February, well below the WHO minimum recommendation of ≥0.2 mg/L at the point of delivery, representing a serious public health concern regarding microbiological safety. Statistically significant spatial variation (P ≤ 0.05) was observed in phosphate concentrations, with elevated levels at distribution-network endpoints (S4–S6, mean 0.238–0.257 mg/L) compared to raw and treated water sites (S1–S2, mean 0.174–0.177 mg/L), suggesting localized contamination within the distribution network. In conclusion, seasonal factors play a dominant role in influencing water quality, while the treatment and distribution system ensures relatively uniform water characteristics across the study area. These findings emphasize the importance of continuous monitoring and optimization of treatment and disinfection processes to ensure safe drinking water quality.

Enhanced Germination and Early Growth of Tacunan Dwarf Coconut (Cocos nucifera L.) through Coconut Water and Auxin Injection

2026-04-13T20:21:39+07:00

Coconut (Cocos nucifera L.) propagation via seednuts is limited by slow and inconsistent germination restricts uniform seedling production in coconut. This study evaluated the effectiveness of coconut water, a natural source of phytohormones, compared with synthetic auxin (NAA) on the germination and early seedling growth of Tacunan Dwarf coconuts under nursery conditions. The experiment was arranged in a randomized complete block design with four treatments: control (no injection), 200 mL NAA + 200 mL distilled water, 200 mL coconut water + 200 mL distilled water, and a combination of NAA + coconut water (100 mL each) + 200 mL distilled water, with four replications. Treatments were applied by injecting the solution near the soft eye at 10 ml/seednut, and germination and growth parameters were recorded over 120 days. Coconut water significantly reduced the time to plumule emergence (61.00 days), produced the longest shoots (142.62 cm), shortened mean germination time (49.52 days), and increased both germination index (0.1527) and seedling vigor index (22,120.44), outperforming both NAA alone and the combined treatment, with a germination rate of 72%. In contrast, NAA resulted in the highest germination percentage (90.67%) but lower seedling vigor, while the combined treatment produced intermediate responses, suggesting no clear synergistic effect between the two stimulants. These findings demonstrate that coconut water can serve as an effective natural alternative for enhancing germination speed and early seedling growth of coconut seedlings.

Design of an Integrated Modern Approach to Detect and Prevent Data Poisoning Attacks in AI Systems: A Multi-Stage Defense Framework for Robust and Secure Learning

2026-04-12T11:10:36+07:00

Adversarial threats like data poisoning attacks affect the training datasets and create biased, worse, or malicious AI models. Static heuristics and high false positive rates have impeded traditional defenses against such adaptive stealthy attacks. These defenses also do not generalize well in federated and non-IID settings. It is difficult to differentiate between poisoned and clean samples, and existing solutions even lack any adaptiveness to newly developed attack strategies; exhibit computational inefficiencies in process. This work presents a Modern Approach to Detect and Prevent Data Poisoning Attacks in AI Systems, coupled with a five-prong mechanism striving for data integrity and model resilience. Those mechanisms are Contrastive Adversarial Poisoning Detector (CAPD), the Federated Memory-Augmented Poison Filter (FMPF), Generative Adversarial Poison Purifier (GAPP), Graph Neural Network Consistency Analyzer (GNNC-A) and Adaptive Meta-Learning Defense System (AML-DS).All integrated, it achieves ≥ 96% poisoning detection accuracy, reduces attack success rates to ≤ 5%, and minimizes false positives to ≤ 2%.This work is scalable and adaptive and provides low computational overhead in security enhancement measures. By showcasing the high integrity of model security without excessively burdening the computer, this work significantly increases the robustness of AI systems in real-world applications.

Reducing GHG Emissions in Frozen Fruit Processing by Solar Optimization and Waste Management Strategies

2026-05-15T18:29:33+07:00

This study quantifies the combined environmental and economic effects of two parallel interventions, solar-panel cleaning and mangosteen peel valorization, at a Thai frozen-fruit processing facility, with emissions calculated using the Thailand Greenhouse Gas Management Organization (TGO) methodology for July–December 2023. Aggregate greenhouse-gas (GHG) emissions fell from 526.34 to 200.12 tCO₂eq/year, a reduction of 326.22 tCO₂eq/year (−62.0%). Waste valorization accounted for 91.4% of the reduction (298.15 tCO₂eq), and solar optimization accounted for the remaining 8.6% (28.07 tCO₂eq). Carbon intensity of production declined from 725.0 to 267.0 kgCO₂eq per tonne of finished product (−63.2%), and the facility BCG (Bio-Circular-Green) composite index rose from 0.42 to 0.78. Solar cleaning increased annual generation from 70,000 to 146,000 kWh and improved the system performance ratio from 0.62 to 0.78 (+25.8%). Peel diversion reached 65.4% (283,120 of 433,120 kg/year), yielding 42,468 kg/year of dried by-products. In economics, capital expenditure totaled 675,000 THB. Annual operating savings reached 549,872 THB (380,000 THB in electricity, 169,872 THB in waste disposal), and by-product sales added 2,831,723 THB/year. Net annual benefit after maintenance was 3,251,595 THB, giving a payback of 2.5 months, a Year-1 ROI of 382%, and a Year-3 ROI of 1,346%. The findings support a phased adoption pathway, solar optimization first, then waste valorization, anchored to source-segregation protocols and seasonal capacity-sharing arrangements among co-located processors.

EffecEffects of Trichanthera gigantea (Ketum Ayam) Leaf Supplementation in Layers Feed on Egg Production and Qualityts of Trichanthera gigantea (Ketum Ayam) Leaf Supplementation in Layers Feed on Egg Production and Quality

2026-03-26T21:42:32+07:00

This research aimed to study the effect of Trichanthera gigantea leaf powder in layer feed on production performance and egg quality. Chopped T. gigantea leaves were dried and powdered. One hundred and twenty layers, aged 6 months, were allotted in a completely randomized design. Birds were divided into four groups according to the level of T. gigantea leaf powder, with 3 replicates per group (10 birds per replicate). Group 1 received feed without T. gigantea leaf powder (control group), while groups 2, 3, and 4 received feed complemented with 1, 2, and 3% of T. gigantea leaf powder, respectively. The layers received feed ad libitum. Feed intake, egg count, and egg weight were recorded. Twenty eggs per group were randomly sampled for egg quality evaluation. The results showed that feed supplementation of T. gigantea leaf powder could significantly improve egg weight, egg percentage, feed conversion ratio, feed per dozen eggs, and egg quality (P<0.01). The study suggests a maximum supplementation of 3%, though 2% is identified as the optimal level for enhancing performance.

Towards Sustainable Electric Vehicle Charging: A Comprehensive Review of Methods, Energy Sources, Charging Technologies, Converter Configurations, and Various Optimization Strategies

2026-04-13T07:02:18+07:00

The changeover from gasoline or petrol engines to electric vehicles (EVs) is motivated by advanced technologies that enhance EV adoption, efficiency, and accessibility. EVs address key challenges by reducing carbon emissions and dependence on fossil fuels. Some key factors, such as EV sources, EV charging methods, EV technologies, converters used in EVs for battery charging, optimization methods incorporated, battery charging times, and EV prices across different sources, should be reviewed to motivate industrialists, researchers, and investors in EVs. This study provides a comprehensive analysis of EV charging technologies, different EV Charging stations (EVCSs), converters, EV charging system structures, and optimization techniques used to improve performance. Renewable energy–based EVCSs utilize locally available energy resources to reduce reliance on fossil fuels. Recent advances in power converter topologies improve power quality and enable flexible AC/DC operation. EV charging methods include wireless charging, battery swapping, and pantograph systems. To meet diverse user demands, EV technologies integrate hybrid systems, fuel cells, batteries, and renewable sources. Optimization techniques enable efficient charging/discharging, grid code compliance, optimal charging station placement, coordinated multi-EV charging, and energy management. This study examines the factors influencing EVs and charging infrastructure, the challenges they face, and future directions, emphasizing the need for coordinated efforts by manufacturers, transport agencies, and governments to support sustainable, large-scale EV adoption.

Tourists’ Responses to Immersive Technology in Community-Based Tourism: A Stimulus–Organism–Response Model

2026-06-11T19:21:33+07:00

Community-based tourism (CBT) has been recognized as an important approach for promoting sustainable tourism, local participation, and community development. However, the role of immersive technology (IMT) in enhancing CBT experiences remains underexplored, particularly how tourists evaluate technology-driven recommendations and respond behaviorally. This study examines the factors influencing tourists’ attitudes and responses to the use of IMT in CBT through the stimulus–organism–response (SOR) model. A cross-sectional survey was conducted in April 2026 with 400 tourists who had experience using IMT within the previous three months. Data were analyzed using partial least squares structural equation modeling (PLS-SEM) with SmartPLS. The results show that recommendation reliability and recommendation accuracy significantly and positively influence tourists’ attitudes toward using IMT, with recommendation accuracy having the strongest effect. In contrast, experiential graphic design does not significantly affect attitude. Furthermore, attitude toward using IMT significantly influences tourist satisfaction and intention to use IMT, but it does not significantly affect intention toward the destination. These findings confirm attitude as an internal psychological mechanism linking technology-related stimuli with tourists’ behavioral responses. The study contributes to the literature by extending the SOR framework to the IMT-based CBT context. It suggests that IMT development for CBT should prioritize accurate, reliable, and useful information rather than focusing only on visual design or technological novelty.

ClothBiz Manager: An Offline GST-Ready Inventory and Billing Automation System Using Rule-Based Domain-Specific OCR

2026-05-01T06:13:52+07:00

Smaller cloth wholesalers in India continue to use manual methods and paper-based systems, leading to discrepancies in inventory and delayed billing, among other issues. This paper describes ClothBiz Manager, an offline inventory management and billing automation solution tailored for Indian wholesale textile businesses. The solution combines a rule-based, domain-specific optical character recognition (OCR) process chain with a normalized PostgreSQL database, implemented using Java Servlets, Java Server Pages (JSP), and the Tess4J library. Our custom-developed Domain-specific Invoice Parser (DIP) processes invoices by applying domain-specific contextual rules and extracts the relevant information, including structured fields such as bale number, LR code, HSN code, and taxes. We evaluated our solution on 150 invoices obtained from Indian textile markets in Tiruppur and Surat. Our method scored 89% on exact matching for structured fields and 78% for multi-line product description fields. Running exclusively on a local server, our application provides absolute data sovereignty and guarantees GST compliance at the database level. Our results clearly show that our system can significantly minimize manual data entry.

Experimental Performance of a Roof Solar Fibre-Cement Chimney and Thermoelectric Power Generation

2026-04-13T20:33:15+07:00

Experimental performance of a roof solar fiber-cement chimney and thermoelectric power generation can enhance ventilation and mitigate heat gain in the attic. This paper presents initial experimental findings on the thermal performance of an advanced roof design using a recently introduced solar chimney configuration that incorporates ten thermoelectric modules. Specifically, one thermoelectric module (MT2-1, 6-127), measuring 40 x 40 mm, was placed in the roof solar fiber-cement chimney, which has a surface area of 0.24 m², and installed on the roof of a small south-facing room measuring 120 cm x 120 cm x 120 cm. Field test results demonstrated that this roof solar fiber-cement chimney, designed with thermoelectric modules, effectively fulfills its intended purposes of promoting ventilation and reducing heat gain through the roof envelope. Therefore, the proposed roof solar fiber-cement chimney with ten thermoelectric modules represents an innovative and viable alternative for various applications, including natural ventilation, indoor air circulation, and reduced roof heat gain. The roof solar fiber-cement chimney and thermoelectric power generation conserve energy and promote solar energy use.

Evaluation of the Role of Silymarin on Certain Biochemical and Histological Parameters in Male Albino Rats Treated with Monosodium Glutamate

2026-06-14T07:10:16+07:00

Monosodium glutamate (MSG), a widely used flavor enhancer, can induce hepatic and renal toxicity through oxidative stress, whereas silymarin from Silybum marianum has antioxidant and hepatoprotective properties. This study evaluated the protective effect of silymarin against MSG-induced hepatic biochemical and histological alterations, with renal involvement assessed through serum biochemical markers, in male albino rats. Thirty male rats (3–4 months) were randomly assigned to three groups (n = 10) and treated for 30 days: T1 (control, 0.9% saline), T2 (MSG, 15 mg/kg body weight), and T3 (MSG 15 mg/kg + silymarin 8 mg/kg). Serum liver enzymes (AST, ALT, ALP), urea, creatinine, and total protein were measured colorimetrically or enzymatically, and liver tissue was processed for histology. MSG significantly (P < 0.05) elevated AST, ALT, ALP, urea, and creatinine, and lowered total protein compared with control; for example, AST rose from 135.2 ± 46.7 to 209.5 ± 64.7 U/L, and creatinine from 1.18 ± 0.37 to 2.67 ± 0.23 µmol/L. Co-treatment with silymarin significantly reversed these changes (AST 152.4 ± 74.5 U/L; creatinine 1.58 ± 0.16 µmol/L; total protein restored toward control), although several values remained different from control. Histologically, MSG caused hepatocyte degeneration, necrosis, and cytoplasmic vacuolation, with loss of the normal radial architecture, whereas silymarin co-treatment largely restored hepatic structure, reducing necrosis and inflammation. These findings indicate that silymarin at 8 mg/kg exerts a meaningful protective effect against MSG-induced hepatorenal toxicity, consistent with its antioxidant and membrane-stabilizing actions.

Synergistic Effect of Antibacterial and Antibiofilm Activity of Oldenlandia corymbosa Combined with Oxytetracycline Against Drug-Resistant Bacteria

2026-03-02T08:16:00+07:00

This study evaluated the antioxidant, antibacterial, synergistic, and antibiofilm activities of Oldenlandia corymbosa, a Thai medicinal plant known for hepatoprotective and anti-breast cancer properties. Antibacterial activity and synergy were assessed using broth microdilution and checkerboard assays, with interactions interpreted by fractional inhibitory concentration index (FICI). Antibiofilm activity was determined using the crystal violet assay. The ethanol extract from aerial parts of O. corymbosa exhibited strong antioxidant activity, with a Ferric Reducing Antioxidant Power (FRAP) value of 5.98 ± 0.12 mg FeSO₄/g extract at 0.1 mg/mL and a DPPH IC₅₀ of 0.19 mg/mL. The extract showed antibacterial activity against 10 of 11 opportunistic bacterial strains, with minimum inhibitory concentrations (MICs) ranging from 20 to 80 mg/mL, except for Proteus mirabilis, which was resistant. In combination studies, O. corymbosa extract demonstrated strong synergistic effects with Oxytetracycline against Staphylococcus aureus (FICI = 0.17). However, no effect was observed against drug-resistant Acinetobacter baumannii or Pseudomonas aeruginosa. Combination with Ampicillin was largely ineffective, except for S. aureus (FICI = 1.016 as an additive effect). The extract significantly inhibited biofilm formation in both standard and drug-resistant strains, including Escherichia coli ATCC 25922, MRSA, P. aeruginosa, and A. baumannii, with inhibition ranging from 64.15% to 75.86% over 4–24 hours. Notably, combinations with Oxytetracycline or Norfloxacin enhanced antibiofilm activity in resistant strains, achieving approximately 68.23%–75.90% inhibition within 8–12 hours. These findings highlight O. corymbosa as a promising multifunctional antimicrobial and antibiofilm agent.

Economic Valuation of Timber and Carbon Stock in the Ban Phru Toei Community Forest, Nakhon Si Thammarat Province, Thailand

2026-03-16T12:03:40+07:00

Community forests provide critical ecosystem services but are frequently undervalued in forest management decisions. This study assessed the economic value of timber resources and quantified the carbon stock of the Ban Phru Toei Community Forest (262 rai; 41.92 ha) in Bang Khan District, Nakhon Si Thammarat Province, Thailand. Twenty-six 20 × 50 m sample plots were systematically established. Tree species richness, density, and stem circumference were recorded; timber volume and above-ground biomass were estimated using published allometric equations. Economic values were derived using the market-price approach, based on 2020 customs import data. A total of 3,812 stems representing 101 species were recorded (1,992 large trees; 1,820 young trees), with a combined density of 146.62 stems per rai. The most abundant species was Madhuca motleyana (1,047 stems), followed by Syzygium aqueum (346 stems) and Stemonurus malaccensis (299 stems). Total timber volume was 2,228.60 m³ (85.72 m³ rai⁻¹). The net timber value of large trees was 25.19 million THB, and the total forest value reached approximately 255.73 million THB. Above-ground biomass in the 26 plots was 24.27 t, yielding a carbon stock of 11.41 t; extrapolated to the entire forest, total carbon was estimated at 114.94 t (0.44 t rai⁻¹), representing above-ground large-tree biomass only and excluding below-ground biomass, litter, dead wood, and soil organic carbon (the dominant carbon pool in peat-swamp forests, estimated at 1,453–3,650 t C ha⁻¹ in comparable Southeast Asian sites). These findings provide baseline data for community-based forest management, conservation planning, and participation in Thailand’s Bio-Circular-Green Economy framework, specifically by enabling future T-VER carbon credit registration and informing spatial zoning decisions for the community forest.

Effects of Cu and Nb Addition on Microstructure and Mechanical Properties of Ultra-low Carbon Steel under Different Annealing Temperatures

2026-03-14T18:07:23+07:00

This paper investigated the beneficial effect of copper (Cu) and niobium (Nb) on grain size and mechanical properties of the ultra-low carbon (ULC) steel annealed at temperatures between 700 and 900 °C. The steel samples were prepared by melting input materials in an induction furnace, followed by hot forging, cold rolling into a thin plate, and annealing under an argon atmosphere. Mechanical testing and microstructural observations showed that the addition of Cu and Nb had a positive effect on the steel’s properties. The grain sizes and the tensile strengths of the steel without Cu and Nb varied in 50-98 μm and 254-204 MPa, respectively, while those of the steel alloyed with Cu and Nb decreased to 40-83 μm and increased to 286-242 MPa at the same annealing temperatures. SEM analysis did not find Cu-enriched zones at the grain boundaries, consistent with the predicted Cu content in this steel. It can be concluded that Nb contributed to grain-size refinement and Cu hardened the ferrite matrix, leading to improved mechanical properties of ULC steel at the same annealing temperatures.

Influence of Chicken Manure, Pig Manure, Cattle Manure, and Coconut Peat on the Growth and Yield of Melon (Cucumis melo L.) cv. Thunder Gold 1497

2026-04-13T20:23:56+07:00

This study investigated the effects of organic growing media mixtures derived from chicken manure, pig manure, cattle manure, and coconut peat on the growth rate, yield quality, and production cost of melon (Cucumis melo L.) cultivar Thunder Gold 1497. The evaluated parameters included the growth rates of plant height, leaf width, leaf number, and stem diameter, as well as fruit sweetness (°Brix), fruit weight, and growing media cost. Mixture ratios were formulated using a mixture experimental design. The results showed that growing media composition significantly affected plant height, leaf width, leaf number, fruit weight, and fruit sweetness (p < 0.05). Cattle manure played a dominant role in enhancing fruit weight and stabilizing plant growth, while pig manure positively influenced fruit sweetness. Optimization using the desirability function indicated that a mixture of 50.00% cattle manure and 50.00% coco peat provided the highest desirability value (0.916), resulting in high growth performance, improved fruit quality, and reduced growing media cost.

Machine Learning Integration in Feature Selection Using Sequential Forward Floating Selection (SFFS) for Improved Credit Scoring

2026-01-19T16:31:48+07:00

Credit risk evaluation serves as the backbone of any financial decision, especially in government funding programs like DOST SET-UP. However, conventional credit assessment methods often face problems such as redundant features, low prediction accuracy, and increased computational complexity, which may result in suboptimal funding decisions. This study presents an integrated machine learning framework that incorporates Sequential Forward Floating Selection (SFFS) to optimize feature selection in credit assessment. This study considered financial and operational parameters from historical data on 33 DOST SET-UP projects undertaken from 2008 to 2016. A total of twelve parameters were initially considered in the feature selection process, which was later reduced to four highly significant features using the SFFS algorithm in Python: the Debt to Asset Ratio (DAR), the Net Profit Margin (NPM), the Liquidity Ratio (LR), and the Return on Investment (ROI). These features were later incorporated into the development of the rule-based fuzzy logic system using the MATLAB platform to produce the final recommendation for the risk assessment of the proposals. The experimental results show that feature selection greatly improved the model's prediction accuracy, and the system achieved an overall accuracy of 94%. The results verified that integrating machine learning-based feature selection and fuzzy logic can develop an efficient and effective credit scoring model for the government's lending programs and improve financial sustainability for micro, small, and medium enterprises.

Phytochemical and Physicochemical Characterization and Heritability Estimates of Fruit Quality Traits in Selected Coconut (Cocos nucifera L.) Populations

2026-04-27T14:39:24+07:00

Limited information exists on the antioxidant performance of coconut cultivars. This study evaluated the phytochemical and physicochemical profiles of fruit quality traits and assessed genetic variability in response to free-radical-induced damage. Data were analyzed using one-way ANOVA at a 95% confidence level, with genetic parameters computed in RStudio v4.2 and graphical representations generated using OriginLab Pro 2024 v10.1. Results indicated that total phenolic content (TPC) was highest in the Tagnanan population (78.54±2.95), followed by hybrid Matag (62.2±2.60). Total flavonoid content (TFC) similarly peaked in Tagnanan (91.83±7.34), with hybrid Matag recording a comparable value (79.42±7.93). DPPH radical inhibition was greatest in Malayan Dwarf and Tagnanan Tall (58.97±1.21 and 58.26±1.00, respectively). Phenotypic coefficients of variation were 26.46% for TPC, 23.28% for TFC, and 5.87% for DPPH. Corresponding genotypic variations were 25.12%, 21.35%, and 4.85%. Heritability estimates were highest for TPC (90.13%) and TFC (84.09%). Genetic advancement as a percentage of the mean (GAM) indicated gains of 49.13% and 40.33% for TPC and TFC, respectively, whereas DPPH scavenging activity showed only 8.25% genetic gain. Physicochemical analysis identified hybrid Matag and Malayan Dwarf as having the highest protein content (16.77% and 16.47%). Hybrid Matag recorded the highest punching force (109.91±0.33 N) and total soluble solids (5.71%), exceeding those of its parental materials. Tagnanan Tall demonstrated superior antioxidant potential, while hybrid Matag combined strong antioxidant traits with excellent physicochemical properties. Both cultivars are recommended for priority inclusion in coconut breeding programs, with multi-location trials advised to validate findings across diverse environments.

Development and Physical Characterization of Molded Charcoal Souvenir Products from Rubberwood Charcoal Reinforced with Tea Waste

2026-03-16T10:10:23+07:00

This research aimed to investigate the physical properties of molded charcoal produced from rubberwood biochar and tea waste, determine the optimal mixing ratio for forming stable products, and evaluate the production cost of the developed souvenir items. Rubberwood charcoal and tea waste were characterized for moisture content, volatile matter, ash content, fixed carbon, iodine adsorption, and moisture adsorption capacity. The rubberwood charcoal exhibited a moisture content of 2.82 ± 0.17%, volatile matter of 85.83 ± 2.07%, ash content of 2.90 ± 0.18%, fixed carbon of 8.45 ± 1.72%, and an iodine adsorption value of 479.94 ± 20.00 mg/g. The tea waste showed a moisture content of 2.75 ± 0.17% and an ash content of 2.83 ± 0.18%. Four formulations with different weight ratios of charcoal powder to tea waste (3:0, 3:1, 1:1, and 1:3) were prepared using tapioca starch as a binder, then formed into pieces using traditional mooncake molds. The 3:1 ratio was found to be the most suitable, providing compact, crack-free specimens with clear surface patterns, no loose powder after drying, and a slightly brown tea color. Moisture adsorption tests over 3 days confirmed that all formulations absorbed moisture, with the 3:1 formulation offering a good balance between structural integrity and functional performance. A preliminary cost analysis indicated that using 1 kg of mixed raw materials yielded 35 molded charcoal pieces at a total cost of 315.84 THB, or 9.02 THB per piece. The results demonstrate the technical feasibility and economic potential of upcycling rubberwood charcoal and tea waste into culturally inspired, environmentally friendly charcoal-based souvenir products with moisture adsorption capability suitable for community-based product development.