-
Research Article
Impact of Electronic Services on Customer Satisfaction: The Case of Commercial Bank in Urban Area of Nepal
Dipendra Bikram Sijapati
,
Manamaya Mishra*,
Bishnu Kumar Khadka
Issue:
Volume 12, Issue 1, June 2026
Pages:
1-10
Received:
25 November 2025
Accepted:
7 January 2026
Published:
29 January 2026
Abstract: Electronic banking (e-banking) plays a crucial role in service delivery for emerging economies like Nepal, yet its effects on customer satisfaction remain uneven and context of dependent. It investigates the influence of major e-banking services are ATMs, internet banking, mobile banking, debit cards, credit cards, and electronic fund transfers on the satisfaction levels among customers of commercial banks in urban areas. This paper based on a descriptive research design, selected respondents 188 bank customers in the Kathmandu Valley by using structured questionnaires to collect their experiences and perceptions. Data analysis drew on descriptive statistics to outline usage patterns, Pearson correlations to examine relationships between variables, and multiple regression to assess predictive impacts while controlling confounding factors. The research findings reveal that ATMs and debit cards exert a significant positive effect on customer satisfaction, likely due to their reliability, accessibility, and ease of use in daily transactions. In contrast, internet banking and mobile banking show no statistically significant impact once other services are accounted for. This stems from persistent challenges like inadequate digital infrastructure, frequent connectivity issues, and usability barriers that hinder broader adoption. Credit cards and electronic fund transfers also fall short of strong influence in this setting. Overall, the regression model accounts for 23.8 percent of the variance in satisfaction scores. These results underscore the importance of dependable transaction mechanisms in driving satisfaction. For emerging economies, the paper recommends investments in robust digital infrastructure, enhanced service quality, and targeted customer education programs to overcome barriers and maximize e-banking benefits.
Abstract: Electronic banking (e-banking) plays a crucial role in service delivery for emerging economies like Nepal, yet its effects on customer satisfaction remain uneven and context of dependent. It investigates the influence of major e-banking services are ATMs, internet banking, mobile banking, debit cards, credit cards, and electronic fund transfers on ...
Show More
-
Research Article
Word Embeddings Network and Transformer Based Part of Speech Tagging for Korean
Pong-Gol You,
Chun-Sik So,
Song-Min Choe,
Yong-Hak Lee,
Chol-Jun O*
Issue:
Volume 12, Issue 1, June 2026
Pages:
11-24
Received:
16 November 2025
Accepted:
9 December 2025
Published:
7 April 2026
Abstract: Korean Part-of-Speech (POS) tagging is different from and more difficult than other languages such as English, Russian and Chinese due to raising issues of Korean word segmentation and analysis of sound-changed morphemes. In this paper we propose a transformer-based Korean POS tagging model, which combines the output vector of a encoder of the transformer with a representational vector of the input word obtained from character-level word embeddings network unlike existing deep learning-based POS tagging models based on BiLSTM. First, in order to perform segmentation of words and changed sound analysis at once, we have designed a model to make a new output sequence of the POS tagging model as a sequence of pairs of strings of morphemes and its POS tags. Second, in order to obtain character-level word representations, word embedding network employing convolution network and highway network are trained. Finally, to make more efficient use of the semantic information of the input word in generating of sequences of POS tagging, we combined the word representation vector obtained from the word-embedding generation network with the output of a encoder of the transformer. According to the experimental results, the proposed model achieves 1.4% performance improvement over the model without incorporating the word representation vector obtained from the word embeddings network, and as a result, the POS tagging accuracy is 96.1%, which is superior to all other compared models including the BiLSTM+CRF model.
Abstract: Korean Part-of-Speech (POS) tagging is different from and more difficult than other languages such as English, Russian and Chinese due to raising issues of Korean word segmentation and analysis of sound-changed morphemes. In this paper we propose a transformer-based Korean POS tagging model, which combines the output vector of a encoder of the tran...
Show More
-
Research Article
The Retail Product Recognition for Intelligent Unmanned Vending Machines Using YOLO and Enhanced Swin Transformer
Issue:
Volume 12, Issue 1, June 2026
Pages:
25-33
Received:
27 November 2025
Accepted:
4 January 2026
Published:
14 April 2026
DOI:
10.11648/j.ajnna.20261201.13
Downloads:
Views:
Abstract: Recently, with the rapid development of deep learning in the retail industry, intelligent unmanned vending machines (UVMs) have become a dominant trend in retail product applications. The most advanced retail product recognition systems for intelligent s are based on object detection models such as Fast R-CNN or YOLO, and image classification models using CNNs like ResNet. However, the complexity of practical environments makes intelligent UVMs face challenges in product recognition performance and many studies are exploring ways to improve this performance. Recently, Vision based Transformers (ViTs) outperform the traditional convolutional neural networks (CNNs) like ResNet, particularly Swin Transformer has gained attention as it is well suited on small dataset and shows promising results compared to the other ViTs. In this paper, we propose two-stage pipeline consisting of product detection and recognition models. The proposed approach utilizes YOLO11 for product detection and an enhanced Swin Transformer for product recognition. Enhanced Swin Transformer is consisted of two main modules: multi-level feature fusion module and a global multi-scale attention module to improve product recognition performance of original Swin Transformer. This proposed approach combines both stages of product detection and recognition into a unified pipeline, leveraging the advantages of each method while avoiding the limitations imposed by single-stage detection models. We demonstrate the effectiveness of the proposed system through the several experiments. The proposed model improves True Acceptance Rate (TAR) by 1.2% when False Acceptance Rate (FAR) is equal to 1e-3. The proposed retail product recognition system using enhanced Swin Transformer demonstrates remarkable generalization capabilities and can handle boundless products without retraining the models.
Abstract: Recently, with the rapid development of deep learning in the retail industry, intelligent unmanned vending machines (UVMs) have become a dominant trend in retail product applications. The most advanced retail product recognition systems for intelligent s are based on object detection models such as Fast R-CNN or YOLO, and image classification model...
Show More
-
Research Article
Neural Networks with Cerebrum Functions Can Exchange Knowledge by Dialog
Seisuke Yanagawa*
Issue:
Volume 12, Issue 1, June 2026
Pages:
34-39
Received:
24 March 2026
Accepted:
7 April 2026
Published:
20 April 2026
DOI:
10.11648/j.ajnna.20261201.14
Downloads:
Views:
Abstract: We can talk about not only the topics in front of us, but also the topics of the past and the future. We are discussing while confirming each other's imagination and understanding. Even in the era when humans did not have letters, they negotiated the exchange of goods, gained new knowledge, and expanded their living area. Of course, using letters further developed civilization to the present day. The group of humans is called society, and the ability to dialogue is essential to be a member of that group. What are the functions required for dialogue ability? This is the starting point of this paper. Trained livestock can understand how to act with the owner's commands or gestures. However, it is not possible to return messages to the owner, such as having leftover work or interruption of work due to inconvenient work etc. This message is the language itself that shows the ability to dialogue. This paper shows that to enable two-way dialogue, both brains need two areas related to language. One area consists of the time series of symbols, and the other area consists of images activated by responses from the symbol area. This structure and behavior are consistent with the knowledge of Cerebrum in neuroscience. The basic logical circuits of neural networks follow Hebb's law in neuroscience, and by connecting them hierarchically, arbitrary logical circuits can be realized with limited scale.
Abstract: We can talk about not only the topics in front of us, but also the topics of the past and the future. We are discussing while confirming each other's imagination and understanding. Even in the era when humans did not have letters, they negotiated the exchange of goods, gained new knowledge, and expanded their living area. Of course, using letters f...
Show More
