Browse
Recent Submissions
Item A hybrid ARIMA and neural network model applied to forecast catch volumes of Selar crumenophthalmus(American Institute of Physics Inc., 2017) Aquino, Ronald L.; Alcantara, Nialle Loui Mar T.; Addawe, Rizavel C.The Selar crumenophthalmus with the English name big-eyed scad fish, locally known as matang-baka, is one of the fishes commonly caught along the waters of La Union, Philippines. The study deals with the forecasting of catch volumes of big-eyed scad fish for commercial consumption. The data used are quarterly caught volumes of big-eyed scad fish from 2002 to first quarter of 2017. This actual data is available from the open stat database published by the Philippine Statistics Authority (PSA)whose task is to collect, compiles, analyzes and publish information concerning different aspects of the Philippine setting. Autoregressive Integrated Moving Average (ARIMA) models, Artificial Neural Network (ANN) model and the Hybrid model consisting of ARIMA and ANN were developed to forecast catch volumes of big-eyed scad fish. Statistical errors such as Mean Absolute Errors (MAE) and Root Mean Square Errors (RMSE) were computed and compared to choose the most suitable model for forecasting the catch volume for the next few quarters. A comparison of the results of each model and corresponding statistical errors reveals that the hybrid model, ARIMA-ANN (2,1,2)(6:3:1), is the most suitable model to forecast the catch volumes of the big-eyed scad fish for the next few quarters.Item Vehicle image classification using data mining techniques(Association for Computing Machinery, 2019-05-24) Suguitan, Agnes S. .; Dacaymat, Lucille N.This paper focuses on the application of different data mining techniques to classify images of vehicles into three classes using Weka. The dataset used for this study were collected from Google Image search engine and other dataset websites. In preprocessing the images, filters such as Color Layout, Edge Histogram and Pyramid Histogram of Oriented Gradients were explored to extract the image features from the dataset. Classification techniques such as Multilayer Perceptron, Sequential Minimal Optimization, Logistic Model Trees, Simple Logistic and Random Forest were used. Results of the study showed that the edge histogram features provided much information to the classifiers in order to correctly classify the images. The SMO classifier performs best with the highest accuracy of 82.37%.Item Effect of weekly changes in environmental parameters on sperm motility characteristics of water buffalo (Bubalus bubalis Linn.) bulls(American Institute of Physics Inc., 2023-06-05) Diaz-Usi, Janice; Venturina, Emma V.; Peralta, Matt Daniel B.; Duran, Peregrino G.; Mingala, Claro N.; Medina, Norraine P.; Baltazar, Eugene; Barroga, Antonio; Rahim, Lellah; Hufana-Duran, DanildaThe effects of climate change on the reproduction of water buffaloes need to be checked in order to employ necessary interventions to maintain or improve the management system and achieve optimal semen quality for the nationwide artificial insemination (AI) program. Fourteen (14) Bulgarian Murrah buffalo bulls were used to determine the effects of weekly changes in ambient temperature, relative humidity, and barometric pressure on the motility characteristics of buffalo sperm cells. Three studies were conducted; 1) Effect of the environmental parameters on motility characteristics of fresh semen, 2) Assessment if the effect of environmental parameters was reflected to the motility characteristics of frozen-thawed spermatozoa, and 3) Assessment of the direct effect of body and scrotal temperature on the motility characteristics of the spermatozoa. In Study 1, correlation analysis showed positive correlation between ambient temperature and motility parameters on distance travelled characteristics and the velocity of movement. Relative humidity had negligible correlation but barometric pressure has significant negative correlation with all the motility parameters except linearity and wobble. In Study 2, the effects of relative humidity and barometric pressure were still distinct in frozen semen but the effect of ambient temperature was negligible. The distance travelled characteristics and velocity of movement including BCF and ALH were negatively affected. THI of 28.5 ºC ambient temperature with 82% relative humidity fall to Moderate level of stress and 32.60 ºC with 70% relative humidity fall to Severe heat stress level showed significant effect on motility characteristics. Analysis on body and scrotal temperatures of the buffalo bulls showed a difference of 0.94 ºC with scrotal temperature lower than body temperature. A significant positive correlation was observed in body and scrotal to ambient temperature and significant negative correlation to barometric pressure. Correlation analysis showed significant positive correlation between testicular temperature and the sperm motility characteristics. Based on the results, it was concluded that in spite of being kept under confinement system of management, high ambient temperature (>28°C), relative humidity (>70%) and low barometric pressure (<1,000 hPA) affect the motility characteristics of water buffalo bulls. Hence, monitoring and intervention to maintain optimal sperm motility characteristics were recommended.Item On lucky primes and their application to Rivest, Shamir, and Adleman (RSA) cryptosystem(American Institute of Physics Inc., 2016-10) Alcantara, N. L. Mar T.; Cabilatazan, Julie Ann Marie S.; Mayugba, Jhun-Jhun A.; Disu, Daisy Ann A.This research study deals with the formulation of an algorithm in finding lucky primes while methodically testing the primality of each. New and different behaviors of lucky primes and related concepts have been explored and exploited to come up with underlying propositions that were supported by proofs made by the researchers. Using these properties, the researchers developed a computer application using the Python programming language in order to generate lucky primes up to certain limit. They tested the viability of using lucky primes as a substitute cryptovariable in a computer program that implements the RSA cryptosystem in encrypting and decrypting messages.Item Usability test of Moodle LMS using empirical data and questionnaire for user interface satisfaction(Association for Computing Machinery, 2022-02) Dilan, Raymund E.; Llavore, Darwin C.; Gali, Mark Anthony C.This study aimed to verify and validate the features of the learning management system used by the Don Mariano Marcos Memorial State University-South La Union Campus, its information and communication technology requirements, and its usability using the QUIS instrument. A total of 1,208 faculty members and students were the respondents in the needs assessment survey and 158 for the usability survey. Adapted survey questionnaires employing Likert-type response scales were utilized to determine the features perceived as important by the end-users and the usability of the LMS. Median, frequency count, and percentage were employed to analyze and interpret the gathered data. Results showed that 38 features were considered important by the end-users of the LMS and that five were not available in Moodle LMS. The ICT requirements were satisfied where the LMS was hosted. The Moodle LMS received a very high level of usability and it is recommended to be implemented in the other campuses of DMMMSU. Furthermore, the Moodle LMS can be further improved by including the features that are not yet present.Item Emotion recognition on selected facial landmarks using supervised learning algorithms(Institute of Electrical and Electronics Engineers Inc., 2018-07) Baculo, Maria Jeseca; Azcarraga, JudithFacial landmarks may be used to localize the movement of facial muscles that help identify an emotion. It is important that these points are appropriately represented to achieve a successful emotion Recognition rate. In this paper, the extraction of 68 facial landmarks, normalization methods and classification of 7 basic emotions are presented. The Cohn-Kanade Database is used as a test bed for the different emotion Recognition tasks. The images are normalized by transforming the inputs based on similarity (CKCT) and the mean shape (CKMS). Forward Search and Principal Component Analysis are used to identify the most important features among the 68 facial points. Decision Tree, Logistic Regression, K-Nearest Neighbor and Multilayer Perceptron algorithms are used in building classifiers on reduced and complete feature set. It is interesting to note that facial points in the mouth area are found to be significant in the classification of emotions. © 2018 IEEE.Item Managing banana bunchy top virus in smallholdings in the Philippines(International Society for Horticultural Science, 2009) Molina A.B.; Sinohin V.G.O.; Dela Cueva F.M.; Esguerra A.V.; Crucido S.S.; Vida E.; Temanel B.E.; Anit E.A.; Eusebio J.E.Banana bunchy top virus (BBTV) is a major constraint to banana production in the Philippines. It severely affects the most popular local cultivar 'Lakatan' (AAA), which is primarily grown on smallholdings by poor farmers. Severe epidemics of BBTV practically eliminated the livelihoods of many such growers in the northern Philippines, who used to supply 'Lakatan' to Manila. The huge local demand for 'Lakatan' has resulted in the establishment of big commercial plantations in the southern Philippines. These enterprises have the resources to manage BBTV by growing virus-free, tissue-cultured plantlets in an annual or 2-year cropping system. In the 1990s, many government-run tissue-culture laboratories were established to produce disease-free planting materials for small-scale growers. However, this programme was not able to realise its aim of bringing an affordable and sustainable supply of clean seedlings to farmers, as they were not cost effective. Also, small-scale farmers were not familiar with the production practices required for tissue-cultured plantlets, and there was a low uptake of the technology. Since then, private tissue-culture laboratories have succeeded in producing 'Lakatan' plantlets at less than half of the cost of those produced by government-run laboratories while ensuring high seedling quality. A rehabilitation programme to restore the livelihoods of small-scale 'Lakatan' growers has begun. Rooted tissue-cultured seedlings produced at participating private tissue-culture laboratories are sent by air to Manila. They are then delivered to the provinces where they are reared to planting stage in village nurseries. Growers are also being trained on improving their banana production system, and the advantages of using disease-free, tissue-cultured planting material, as opposed to suckers, are being demonstrated. Although plants derived from tissue culture became re-infected in the field, the symptoms appeared late and usually occurred in the ratoon crop. In heavily infested areas, the value of annual cropping was evident. This work is helping in the formulation of a sustainable 'Lakatan' production system for small-scale farmers.Item Spatio-temporal urban land use/cover change analysis in a hill station:(Elsevier Ltd, 2011) Estoque, Ronald C.; Murayama, YujiThis study explores the spatial and temporal characteristics of land use/cover (LUC) changes in Baguio city, the only American hill station in Asia and the summer capital of the Philippines. Remote sensing (RS) satellite images were used to develop the 1988, 1998, and 2009 LUC maps of the city in a Geographic Information Systems (GIS) platform. Results reveal that the city has undergone a major physical landscape transformation for the last 21 years as indicated by a rapid built-up area expansion and substantial changes in its other land uses/covers. This study also analyzes the spatial pattern of urban growth in Baguio city. Furthermore, it presents insights in planning for the future sustainable urbanization of this highly valued city. © 2011 Published by Elsevier Ltd.Item Analytic hierarchy process in geospatial analysis(Analytic hierarchy process in geospatial analysis, 2012-09) Estoque, Ronald COwing to the increasing demand for land, forests, waterways, and other depleting resources brought about by increasing population growth, there is an urgent need for a scheme that can promote their sustainable utilization. The best use of these resources must be selected so that they remain available to the generations to come. Natural disasters like earthquakes, landslides, and floods have been major concerns in many countries. This makes it necessary for planners to design sound risk-management contingencies to prepare for such disasters. Most of the time, however, decision makers have different and conflicting priorities, concerns, knowledge, and expertise in dealing with these problems. This reality complicates the decision-making process on how a particular resource should be utilized, and on how an analysis of the susceptibility of a particular area to a certain disaster risk hazard should be carried out. In recognition of the complexity, magnitude, and importance of these problems, a decision-making technique that is responsive, transparent, and acceptable to the decision makers and other stakeholders is needed. Multi-criteria decision making (MCDM), or multi-criteria decision analysis (MCDA), is a decision-making technique that helps decision makers who are confronted with conflicting priorities to come up with an acceptable decision using a transparent decision-making process. It has been one of the fastest growing problem-resolving approaches in the past decades (Triantaphyllou 2000).Item Agroforestry practices in a critical watershed(Journal of Agriculture and Rural Development in the Tropics and Subtropics, Supplement, 2005-08) Salamanca, Eric F.; Muuss, UweAgroforestry involves the systematic integration of tree crops and agricultural/food crops under one management scheme. This land-use system is practiced in the hillsides of the tropics but thought in the past to be inappropriate in critical watersheds. This paper aims to present various agroforestry practices in the Kaliwa watershed in the Philippines which is considered as critical. It is located in the municipalities of Tanay and Gen. Nakar in the provinces ofRizal and Quezon, respectively. It covers an area of 27,608 hectares. The area is characterized by various land uses, landscapes, vegetation and rivers. About 7000 poor rural households or approximately 37,868 individuals in the eight villages inhabit inside in the watershed. To improve the ecological well being of the watershed while enhancing the socio-economic status of the forest occupants, the government adopted a land-use system called agroforestry. It focuses on the integration of forest tree crops, agricultural crops, small-ruminants, native chicken, fish and other components in one management system. The land use system is designed to allow the production of crops to meet their immediate needs of a family without sacrificing environmental sustainability. Various agroforestry farming systems were set up in different villages. Nitrogen fixing tree species, botanical pesticide tree, vetiver grass (for erosion control) horticultural fruit trees were planted. The alleys were planted with mungbean, soybean, pineapple rice and corn. Goat and sheep were raised using the concept of cut-and-carry system. Native chicken are raised to provide egg and meat for the farmers. The integration of fish production and tree growing are established Nitrogen trees were planted around the aqua-silviculture pond. Other crops such as Tiger grass are planted in the alley to reduce soil erosion and the flowers can be made into soft brooms, thus augmenting farm income while minimizing soil erosion. All organic materials such as litter fall, trimmings, weeds and the like are chopped/shredded and decomposed with the help of native surface-earthworms. The vermin-composts were applied to the vegetables or other crops. The agroforestry practices that were introduced in the watershed are adopted by the farmers with great enthusiasm.Item Kitikit: A mobile app for searching hand-carved wood products(Association for Computing Machinery, 2019) Balcita, Amy P.; Kupsch, Stephan; Palaoag, Thelma D.While the use of mobile apps has been very successful in developed countries around the globe, in Pugo La Union, Philippines especially in the wood carving industry, it does not have a significant presence. This research was conceptualized for the benefit of the woodcarving industry stakeholders, i.e., the wood carvers, market vendors, distributors, wholesalers, shop owners, and the government. Through this study, the researchers made an effort to examine the problems encountered by the woodcarving business owners and put forth a desirable recommendation, the design of a mobile application that is beneficial to the woodcarving industry. The researchers developed the KITIKIT, a mobile application to ease the searching for hand-carved wood products. KITIKIT has integrated Google Maps to allow users to trace the location of an establishment selling hand-carved wood products. Also, KITIKIT provides a facility for establishment owners to advertise by enabling them to upload pictures of their establishments and products, and provide relevant information such as price, contact details, customizations options. It is very significant for the woodcarving industry to develop and offer mobile apps to satisfy the purpose and needs of customers. Future researchers are encouraged to conduct a usability testing of the proposed mobile application. © 2019 Association for Computing Machinery.Item Image-based classification and segmentation of healthy and defective mangoes(SPIE - The International Society for Optical Engineering, 2019) Baculo, Maria Jeseca C.; Ruiz, ConradoThe use of image processing and classification for agricultural applications has been widely studied and has led to work such as the automatic grading of fruit and vegetables, yield approximation and defect detection. Image segmentation is one of the first steps to identify the region of interest within an image. This paper presents an approach to automatic segmentation and classification of healthy and defective Carabao mangoes. K-means, range filtering and color-channel segmentation were utilized so that the varying texture and color of mangoes due to the surface defects can be considered. Results show that the proposed technique performs better than the classical K-means segmentation. The performance of segmentation step has a considerable influence on the precision of the classification model. Segmented and not segmented images were trained using KNN, SVM, MLP and CNN. The experiments showed that the models performed better when trained with segmented images.Item Deep learning-assisted milkfish fingerling counting:(Association for Computing Machinery, 2024) Sibayan, Christian Jose; Abando, Benjie R.; Acosta, Justin P.; Madrelijos, Johnico D.; Marzan, Charlie S.The manual counting of milkfish fingerlings within the fish industry in the Philippines poses significant challenges due to its time-consuming nature, labor intensiveness, and susceptibility to errors. To address these issues, this study presents a deep learning approach for the real-time counting of milkfish fingerlings to enhance both the speed and accuracy of fingerling counts. The dataset was meticulously curated from six milkfish fingerlings farms, with gathered videos converted into image frames and subjected to polygon annotation. Furthermore, various augmentation techniques were applied to these images to enhance training efficacy before integration into a deep neural network. Notably, implementing the YOLOv8n-seg object detector with segmentation and counting yielded a remarkable accuracy rate of up to 90.64% and 93.34%, respectively. This innovative methodology promises significant advancements in the automation and efficiency of milkfish fingerling counting processes within the fish industry, paving the way for improved productivity and accuracy in fishery management practices. © 2024 ACM.Item Next bar predictor:(Proceedings of the 2020 IEEE International Conference on Communication and Signal Processing, ICCSP 2020, 2020-07) Dungan, Belinda M.; Fernandez, Proceso L.Music generation has been an active field of research in computer science and is considered as a creative task attempting to imitate human creativity. With the different approaches to generate musical content, recent works have focused on general adversarial networks. One of these is the Midinet, which is considered the baseline model in this study. In this paper, we propose our Next Bar Predictor, a generative model that creates melody one bar at a time using the previous bar as basis to generate aesthetically pleasing melodies. We explore several variants of this by experimenting on different regression and classification models such as Decision Trees (DT), K-Nearest Neighbors (KNN), and Multilayer Perceptron (MLP). The models were trained using the dataset from Theorytab which consists of 460 songs. The outputs of these different variant models were then compared against those from the Midinet, using both machine-based objective scoring mechanism as well as human-based subjective evaluations. The dissimilarity scores obtained by our KNN (0.65) and DT (0.74) models, scored against the melodies in the dataset, are sufficiently high and indicates that both models are generally creative. Furthermore, based on the evaluation by human listeners, the melodies generated by our DT models are more realistic and pleasing than those of the Midinet. Casual listeners also prefer the DT model to be more interesting, although professional listeners think otherwise. Finally, all the variant models, when compared with Midinet, require much less training time and computational power. The proposed Next Bar Predictor is therefore a viable alternative for automated music generation.Item Automatic mango detection using Image processing and HOG-SVM(Association for Computing Machinery, 2018-12) Baculo, Maria Jeseca C.; Marcos, NelsonMango is an agricultural produce with high export value as it is being consumed internationally. To ensure its production yield, the manual handling and classification tasks should be performed with precision and care by local farmers. Image processing and machine learning has improved the way classification, defect detection, and yield approximation are handled. Detection is considered as an initial step prior to performing these tasks. This paper presents an automatic mango detector by combining a Support Vector Machine (SVM) classifier trained with Histogram of Oriented Gradients (HOG) features and image segmentation. The image segmentation performed on both HSV and RGB color spaces using image processing techniques achieved a mean IoU of 0.7938. A HOG-SVM based classifier was trained and achieved an F-score of 89.38%. Results show that combining segmentation with HOG-SVM can detect and localize healthy and defective mango images with different background color and illumination.Item Disaster sentiment analysis: addressing the challenges of decision-makers in visualizing netizen tweets(Institute of Physics Publishing, 2020) Baro, R. A.; Palaoag, T. D.Among other crucial aspects of disaster-related initiatives is decision-making. The capacity to carry out efficient holistic management of measures and programs rely greatly upon the decision-makers and the religious participation of all stakeholders. Adopting the sentiments can help government leaders in their decision-making responsibilities towards disaster management when meaningful patterns are easily visualized. The purpose of this research is to analyze disaster-related sentiments from Twitter and presents the results using data visualization easily understood by decision-makers. The study underscores the use of Python programming with NLP techniques to learn from Twitter data, Sentiment Analysis using TextBlob to identify polarity and subjectivity of tweets and Plotly for interactive data visualization. The implementation of the study contributes to the reduction of injury, damage to infrastructures, properties and especially the loss of life. The challenges of both decision-makers and data scientists working in the discipline are highlighted in the study. © Published under licence by IOP Publishing Ltd.Item Applying feedback analysis on citizen's participation system of CSFLU barangays on disaster preparedness(Institute of Physics Publishing, 2018) Ocampo, A.J.; Baro, R.; Palaoag, T.Various initiatives through the use of ICT paved the way to better improve the services of the government during disaster situations. It helped in the preparation and mitigation process during disaster situations through different mediums such as Social Networking Sites and SMS to disseminate information. However, data that are gathered from this medium are not sufficient to address the problem experienced by the citizens, thus the concept of Citizen's participation system was developed. The objective of the study is to provide a mechanism or tool for barangay officials and the city government to strategically plan preventive measures during times of disasters based on the citizen's perspective, data analytics gathered from sentiments, suggestions, and feedback of the citizens was analysed using of Feedback Analysis in order to provide accuracy of data which is needed by the disaster response team that will be generated through data analytics. © Published under licence by IOP Publishing Ltd.Item A survey of AI-based detection techniques for intracranial aneurysms(Institute of Electrical and Electronics Engineers, 2026) Dacanay, Kathrina T.; Ingosan, Jeffrey S.Early detection of intracranial aneurysms (IA) is essential for preventing subarachnoid hemorrhage, a condition associated with high morbidity and mortality. Recent advances in artificial intelligence (AI), particularly deep learning, have shown promising results in assisting radiologists across different angiographic modalities. This survey synthesizes findings from 23 peer-reviewed studies spanning CTA, TOF-MRA, DSA, 3D-RA, and multiphase CTA. Across modalities, CTA-based models achieved some of the highest reported patient-level sensitivities, often reaching ${95-100 \%}$, while multiphase CTA further improved sensitivity by $4-6 \%$ compared to single-phase scanning. TOF-MRA studies reported sensitivities ranging from $90-96 \%$, with segmentation models attaining Dice scores of $0.80-0.83$ and AUC values up to 0.96. DSA and 3D-RA systems demonstrated strong performance for larger aneurysms, achieving sensitivities of $91-97 \%$, though performance dropped for aneurysms $\leq 3$ mm. Across all modalities, 3D CNNs and U-Net variants were the most frequently used architectures, with two-stage detection-segmentation pipelines consistently reducing false positives. Despite these advancements, challenges remain, particularly false-positive rates in CTA, reduced sensitivity for very small aneurysms, and limited cross-institutional generalization. This review consolidates current trends, highlights persistent gaps, and outlines future directions for developing clinically robust AI systems for IA detection.Item Exploring ICT competencies and innovation readiness of school librarians:(International Research and Innovation Journal on Education, Engineering and Agriculture, 2025-06) Reyes, Alma F.; Siago, Sonia B.The important and dynamic role of school librarians, especially with the rise of digital technology requires that they continually excel in competencies in both library and information sciences. This descriptive-correlational study assessed the ICT competencies of school librarians in the 1st District of Pangasinan using the National Competency-Based Standards for Filipino Librarians. It focused on core areas such as email, hardware, internet, operating systems, software applications, web tools, and the application of information technologies. The study also examined the challenges librarians face in applying these skills and the relationship between their competence levels and the difficulty of these challenges. Findings revealed that librarians showed high competence, especially in email and operating systems, while challenges were generally rated as slightly difficult. Statistical analysis indicated no significant relationship between ICT competence and the degree of difficulty, suggesting that higher skill levels do not necessarily equate to fewer challenges. As a result, a Strategic ICT Capacity Enhancement Plan was developed to further improve competencies and address the difficulties school librarians encounter in applying ICT skills.Item Apiculturists’ issues and challenges:(IOP Conference Series: Materials Science and Engineering, 2019) Kupsch, Stephan; Palaoag, Thelma D.; Balcita, AmyThe most significant activity of honey bees, as far as advantages to people, is their pollination of natural vegetation. Bees and other pollinators seem to be declining globally. Beekeeping in the Philippines is a thriving industry that perfectly matches the natural landscape of the country. The goal of this study was to define the challenges and issues of Apiculturists in the Philippines where-in an IT approach was formulated to address these issues. Up to now, there is no initiative Information Technology approach in the beekeeping industry in the Philippines. Design thinking has been used as methodology in this study. Several challenges and issues were determined, and it is stated that the main problem to be addressed in order for the bee colonies to have a longer lifespan is its Apiary Location, an IT solution via Drone Technology with a Computer Vision approach was proposed named as Apiary Locator. A proposed framework of the Apiary Locator has been prepared. It is then recommended for the Apiary Locator to be developed as it shall benefit the apiculturists of the Philippines. Also, it will be a basis for developing IT solutions for the Beekeeping industry.