Fit-hcmus

KHOA CÔNG NGHỆ THÔNG TIN

TRƯỜNG ĐH KHOA HỌC TỰ NHIÊN (ĐHQG-HCM)

Giới thiệu

ƯƠM MẦM TINH HOAVỮNG BƯỚC VƯƠN XA

Trang kỷ niệm

HOẠT ĐỘNG HỢP TÁCDOANH NGHIỆP

Xem chi tiết

LỊCH LÀM VIỆC CỐ VẤN HỌC TẬP

Nhằm hướng dẫn, hỗ trợ, giải đáp sinh viên trong quá trình học tập, xây dựng kế hoạch và có phương pháp học tập hiệu quả.

Xem chi tiết

~30

Năm

Chào mừng đến với..

Khoa Công nghệ thông tin

Khoa Công nghệ thông tin được thành lập theo quyết định số 3818/GD-ĐT ngày 13/12/1994 của Bộ Trưởng Bộ GD&ĐT, dựa trên Bộ môn Tin học (thuộc Khoa Toán, Trường Đại học Tổng hợp TP.HCM). Trải qua hơn 28 năm hoạt động, Khoa đã phát triển vững chắc và được Chính phủ bảo trợ để trở thành một trong những khoa Công nghệ thông tin hàng đầu trong hệ thống giáo dục đại học của Việt Nam.

4500+

Sinh viên đại học

500+

Học viên cao học

100+

Giảng viên

Tin tức & Sự kiện

Thông báo sau đại học

Thầy Cao Xuân Nam bảo vệ thành công Luận án Tiến sĩ ngành Khoa học máy tính

• 11/02/2026

• 
• 

Tốt Nghiệp > Qui trình thực hiện

[CQ] DSSV chính thức thực hiện đề tài tốt nghiệp Khóa 2022-đợt 2 (bảo vệ T7/2026)

• 10/02/2026

• 

Tốt Nghiệp > Qui trình thực hiện

Thông báo nộp đơn đăng ký bảo vệ/ huỷ bảo vệ đề tài tốt nghiệp khóa 2022, đợt 1 (tháng 3/2026)

• 09/02/2026

• 
• 

Thông báo sau đại học

[SĐH] Mời tham dự buổi bảo vệ LATS cấp CSĐT của NCS Cao Xuân Nam (8g00, 10/02, phòng F102)

• 06/02/2026

• 
• 

Thông báo sau đại học

[SĐH] Kế hoạch bảo vệ đề tài tốt nghiệp dành cho học viên cao học năm 2026

• 02/02/2026

• 
• 

Thông báo liên thông môn học ĐH-ThS

Thông báo xét tuyển sinh viên tham gia chương trình liên thông môn học ĐH-ThS năm 2026

• 02/02/2026

• 

Thông báo chung

Thư ngỏ mời Doanh nghiệp tham gia Ngày hội việc làm 2026

• 26/01/2026

• 
• 
• 

Tốt Nghiệp > Qui trình thực hiện

[CQ] Điều chỉnh thời gian bảo vệ các đề tài tốt nghiệp Khóa 2022-1 (tháng 3/2026)

• 26/01/2026

• 
• 

Hội thảo - Hội nghị

Tổng kết Seminar giới thiệu Học bổng Thạc Sỹ Fulbright toàn phần tại Hoa Kỳ năm học 2027 - 2028

• 22/01/2026

• 
• 

Hội thảo - Hội nghị

Tổng kết Buổi giới thiệu về các chương trình PhD và hoạt động nghiên cứu của Nanyang Technological University (NTU), Singapore

• 17/01/2026

• 
• 

Thành tích sinh viên

Kết quả đạt giải Cuộc thi Olympic Tin học Sinh viên trường Đại học Khoa học tự nhiên năm 2025

• 16/01/2026

• 
• 
• 

Thông báo sau đại học

Thông báo về việc đăng ký chuyển sang hệ tự túc áp dụng cho học viên cao học khoá 33/2023

• 13/01/2026

• 
• 

Thông báo chung

Cập nhật thành tích sinh viên khoa CNTT HK1/ 2025 - 2026

• 12/01/2026

Thông báo chung

Hội nghị khoa học trẻ Trường Đại học Khoa học tự nhiên, ĐHQG-HCM Lần 3

• 12/01/2026

• 

Hội thảo - Hội nghị

Đăng ký tham gia Buổi giới thiệu về các chương trình PhD và hoạt động nghiên cứu của Nanyang Technological University (NTU), Singapore

• 12/01/2026

• 
• 

Chương trình đào tạo

Đại học chính quy

Chương trình chuẩn

Chương trình đào tạo cử nhân của Khoa Công nghệ Thông tin đã được đánh giá theo bộ tiêu chuẩn AUN-QA và được đánh giá cao nhất cả nước trong đợt đánh giá ngoài tháng 12/2009.

Tìm hiểu thêm

Chương trình tiên tiến

Chương trình đào tạo ngành Khoa học máy tính và giảng dạy hoàn toàn bằng tiếng Anh. Chương trình đang là sự lựa chọn ưu tiên của nhiều sinh viên có thành tích học tập xuất sắc ở bậc phổ thông, hoặc tại các cuộc thi học thuật trong nước và quốc tế.

Tìm hiểu thêm

Chất lượng cao

Chuẩn đầu ra của chương trình được xây dựng theo cách tiếp cận CDIO, đảm bảo người học được trang bị đầy đủ kiến thức, kỹ năng và thái độ, đáp ứng nhu cầu xã hội khi tốt nghiệp. Chương trình được xây dựng cân đối dựa trên việc trang bị vững vàng kiến thức nghề nghiệp và phát triển kỹ năng cá nhân, kỹ năng mềm và ngoại ngữ cho sinh viên.

Tìm hiểu thêm

Chương trình khác

Ngoài ra, Khoa CNTT còn có các chương trình khác như: Chương trình cử nhân liên kết với ĐH Claude Bernard Lyon 1 (Việt - Pháp), Đào tạo từ xa qua mạng...

Tìm hiểu thêm

Đào tạo Sau đại học

Khoa học máy tính

Chương trình cung cấp những kiến thức và kỹ năng chuyên sâu thuộc lĩnh vực Khoa học Máy tính như Trí tuệ nhân tạo, Khoa học dữ liệu, Máy học, Thị giác máy tính, Xử lý ngôn ngữ tự nhiên, An ninh thông tin, và Vạn vật kết nối (IoT). Học viên còn được đào tạo các kỹ năng cá nhân, kỹ năng nhóm, kỹ năng quản lý, phương pháp nghiên cứu khoa học cũng như cách thức triển khai và xây dựng các hệ thống thông minh hiệu quả.

Hệ thống thông tin

Chương trình cung cấp những kiến thức và kỹ năng chuyên sâu thuộc lĩnh vực hệ thống thông tin (HTTT), đặc biệt là các chủ đề nâng cao liên quan đến kiến trúc tổng thể HTTT của tổ chức. Chương trình đào tạo các chuyên gia, các kỹ sư, các phân tích viên có tầm nhìn sâu và rộng nhằm giúp các tổ chức, cơ quan xây dựng và thực thi chiến lược phát triển và ứng dụng HTTT của mình.

Trí tuệ nhân tạo

Chương trình cung cấp những kiến thức và kỹ năng chuyên sâu thuộc lĩnh vực trí tuệ nhân tạo (TTNT). Khác với ngành Khoa học máy tính, chương trình ngành TTNT tập trung chuyên sâu về TTNT với các môn học như TTNT nâng cao, Học máy nâng cao, Khai thác dữ liệu lớn, Học máy với dữ liệu đồ thị, và TTNT trên vạn vật.

Tìm hiểu thêm

Khoa học máy tính

Hệ thống thông tin

Trí tuệ nhân tạo

Thông tin tuyển sinh

Nghiên cứu

Một số kết quả nổi bật

Khoa Công nghệ thông tin tập hợp một đội ngũ nghiên cứu mạnh với các chuyên gia có kinh nghiệm trong nghiên cứu cơ bản cũng như phát triển các giải pháp ứng dụng trí tuệ nhân tạo, máy học, dữ liệu lớn, tối ưu hoá, tương tác người máy, thị giác máy tính, xử lý ngôn ngữ, và công nghệ phần mềm.

Tìm hiểu thêm

2025

LTDAD-Talker: Landmark-Guided Talking Face Generation with Temporal Consistency and Detail-Aware Discriminator

Generating talking face videos from audio has gained significant research interest due to its various applications. However, end-to-end approaches that simultaneously reconstruct facial features and synchronize lip movements often struggle to balance these tasks effectively, resulting in low video quality with blurred details and inconsistent identity preservation as well as disconnected facial movements throughout videos. To address these challenges, we propose a novel two-stage framework (LTDAD-Talker) that separates the process into an Audio2Lmk module for generating precise facial landmark sequences and a Lmk2Lip module that creates high-quality facial imagery guided by these...

Các tác giả

• Duc Khoan Le
• Duc Hao Do
• Minh Hoang Pham

2025

Enhancing Sequential Recommendation Systems through Knowledge Graphs

Sequential recommendation models have achieved strong performance by capturing users’ temporal preferences from historical interaction sequences. However, they often lack the ability to leverage rich semantic information about items, limiting both their accuracy and explainability. In this work, we propose a novel framework that enhances sequential recommendation by integrating a Knowledge Graph-based Intent Network (KGIN) through a late-fusion strategy. KGIN models latent user intents as attentive combinations of knowledge graph relations and performs relational path-aware aggregation to encode long-range semantic dependencies into user and item...

Các tác giả

• Luong Thanh Tu
• Tran Quoc Anh
• Tran Duy Hoang
• Nguyen Ngoc Minh Chau
• Trieu Nhat Minh

2025

CrossPAR: Enhancing Pedestrian AttributeRecognition with Vision-Language Fusion and Human-Centric Pre-training

Pedestrian attribute recognition (PAR) is crucial in various applications like surveillance and urban planning. Accurately identifying attributes in diverse and intricate urban environments is challenging despite its significance. This paper introduces a novel network for PAR that integrates a human-centric encoder, trained on extensive human datasets, with a vision-language encoder, trained on substantial text-image pair datasets. We also develop a cross-attention mechanism utilizing a Mixture-of-Experts approach that combines the human-centric encoder's proficiency in local attribute detection with the vision-language encoder's ability to comprehend global...

Các tác giả

• Bach-Hoang Ngo
• Si-Tri Ngo
• Phu-Duc Le
• Quang-Minh Phan
• Minh-Triet Tran
• Trung-Nghia Le

2025

SECURITY POLICY MANAGEMENT IN UNIFIED ATTRIBUTEBASED ACCESS CONTROL MODEL

This paper presents UABAC, a Unified Attribute-Based Access Control framework that aims to flexibly and scalarly integrate traditional access control models, including Discretionary Access Control (DAC), Mandatory Access Control (MAC), and Role-Based Access Control (RBAC). We propose a systematic mapping of the fundamental principles of these traditional models into ABAC, ensuring semantic equivalence while maintaining expressiveness. In addition, we introduce the cover of the policy set and inference rules concepts which set the foundation for managing access control rules within...

Các tác giả

• Pham Thi Bach Hue
• Minh-Triet Tran

2025

Heuristic algorithms for decoding phase of the non-adaptive group testing

Heuristic algorithms for decoding phase of non-adaptive group testing Abstract: Group testing is a longstanding problem with numerous applications in social life, it plays an important role in quickly performing sample tests and identifying and localizing the defect items on a large scale, in conditions where the number of tests is limited. The step of analyzing the pooling test matrix to predict the desired results requires certain technical probability calculations and combinatorial...

Các tác giả

• Le Phuc Lu
• Vu Tien Luc
• Thoi Gia Nghi
• Xin Quy Hung

2025

Attention-Driven with Gaussian Processes for Weakly Supervised Hemorrhage Detection in Brain CT Scans

Intracranial hemorrhage (ICH), a life-threatening emergency, requires rapid diagnosis through analysis of computed tomography (CT) scans. Automating intracranial hemorrhage detection using artificial intelligence is crucial, but is hampered by the scarcity of labeled training data. To overcome this limitation, we present a novel weakly supervised deep learning framework. This framework integrates Attention-based Multiple Instance Learning (Att-MIL) with Sparse Variational Gaussian Processes (SVGP) to improve diagnostic accuracy and...

Các tác giả

• Huynh Si Kha
• Trieu Nhat Minh
• Tran Duy Hoang

2025

AcaQAS: An Academic Question Answering System Based on Finetuning Large Language Models

Recent advancements in large language models (LLMs) have demonstrated their effectiveness across various natural language processing tasks, including question answering. However, adapting these models to domain-specific applications, such as academic question answering (Academic QA), presents challenges, particularly when computational resources are limited. This paper explores the use of parameter-efficient fine-tuning techniques, specifically Low-Rank Adaptation (LoRA), to optimize open-source LLMs for academic QA...

Các tác giả

• Thai Hoang Le
• Bao Thai Duong

2025

MESN: A multimodal knowledge graph embedding framework with expert fusion and relational attention

Knowledge graph embedding is essential for knowledge graph completion and downstream applications. However, in multimodal knowledge graphs, this task is particularly challenging due to incomplete and noisy multimodal data, which often fails to capture semantic relationships between entities. While existing methods attempt to integrate multimodal features, they frequently overlook relational semantics and cross-modal dependencies, leading to suboptimal entity representations. To address these...

Các tác giả

• Tran Huy Ban
• Le Ngoc Thanh

2025

EXPLAINING SEMANTIC ROLE LABELLING OUTPUTS WITH WORD FEATURE IMPORTANCE

Mặc dù các mô hình học sâu mang lại hiệu quả dự đoán ấn tượng trong các tác vụ xử lý ngôn ngữ tự nhiên (NLP), chúng cũng dấy lên lo ngại về độ tin cậy và những vấn đề đạo đức bởi tính chất hộp đen của chúng. Vì vậy, một bài toán mới đang thu hút nhiều nỗ lực nghiên cứu là giải thích mô hình NLP thông qua việc làm rõ tầm quan trọng của các đặc trưng đầu vào đối với dự đoán đầu ra của các mô hình học sâu...

Các tác giả

• Tuan Nguyen Hoai Duc

2025

Struct-KGS2S: a structural context-based sequence-to-sequence model for link prediction in knowledge graphs

Nowadays, description-based approaches are widely employed for the link prediction task on knowledge graphs, largely due to the integration of pre-trained language models (PLMs). These approaches enable the effective exploration and utilization of the rich textual data present in knowledge graphs. However, despite their strengths, models that rely on textual feature descriptions still face certain limitations compared to structure-based methods, particularly in terms of providing additional contextual information for...

Các tác giả

• Hien Dang
• Thu Nguyen

2025

RE-VC: Robust zero-shot voice conversion model for realistic environments

Voice conversion (VC) technology, which enables the transformation of one speaker’s voice to that of another while preserving linguistic content, has seen significant advancements in recent years. However, existing VC models face two major challenges: reduced voice quality when processing noisy speech and diminished voice similarity in zero-shot inference scenarios. In this paper, we propose a novel approach to address these issues by integrating a content encoder with WavLM model to effectively extract linguistic features and an attention-based speaker encoder to capture speaker-specific...

Các tác giả

• Ho Cong Luong
• Do Duc Hao
• Nguyen Hai Minh
• Chau Thanh Duc

2025

Multi-Modality Fusion for Enhanced Driving Scene Semantic Segmentation

Semantic segmentation in autonomous driving scenes is essential for precise environment perception and decision-making processes. While single modality systems demonstrate efficacy for specific applications, they frequently fall short in addressing the complexities of diverse data. Multi-modal systems thereby overcome these limitations by integrating multiple data types, providing a more comprehensive and accurate analysis. In this paper, we propose an approach to enhance semantic segmentation performance through multi-modality data...

Các tác giả

• Ngoc-Linh Nguyen-Ha
• Gia-Khanh Le
• Xuan-Tung Nguyen
• Hai-Dang Nguyen
• Ngoc-Thao Nguyen

2025

Integrating Local and Global Features for Enhancing ViVQA model

Generative AI enables the creation of synthetic data to train AI models, offering new possibilities for expanding training datasets and overcoming data limitations. However, generating images for Visual Question Answering (VQA) is challenging, as the augmented images must preserve the semantic consistency of the <image, question, answer> triplet. This study investigates the use of image generation models to augment data and enhance VQA...

Các tác giả

• Nguyen Huynh Phu Thinh
• Nguyen Tien Huy
• Le Thanh Tung

2025

Combined Approaches for Collaborative filtering Recommender systems

The rapid growth of digital data has made recommender systems essential for delivering relevant information across various domains. This work aims to enhance recommendation accuracy by combining the strengths of two widely used approaches in collaborative filtering: neighbor-based models and latent factor models. Neighbor-based models provide high interpretability through user similarity, while latent factor models capture deep user-item interactions for more accurate...

Các tác giả

• Le Nguyen Hoai Nam

2025

Enriching Information for Speech Emotion Feature Using Polynomial Chirplet Transform and Density-Based Clustering Algorithm

This paper addresses the problem of speaker emotion recognition. Modeling various emotions is challenging due to their complex and non-linear characteristics. The Polynomial Chirplet Transform (PCT) is particularly well-suited for capturing such emotional variations in speech signals. This paper designs a new algorithm for estimating the optimal polynomial functions for PCT when analyzing single signals. Since a dataset generates numerous polynomial...

Các tác giả

• Do Duc Hao
• Chau Thanh Duc
• Tran Thai Son

2025

A video recommendation system based on a combination of user experiences and video content

Recommendation systems are crucial for personalizing content on video platforms. A main approach in video recommendation is item-based collaborative filtering, which predicts user interest by analyzing ratings of similar videos and aggregating these ratings using weighted influence factors. However, these systems face challenges in accurately computing video similarity and estimating influence between videos, as many existing methods rely solely on user experience...

Các tác giả

• Do Thi Thanh Ha
• Le Nguyen Hoai Nam

2024

Support Learning Design and Analytics with EduP Knowledge Model

Learning design (LD) have been a prominent topic in the academic community for many years. It aims at planning and organizing learning activities and resources to promote learning process and engage students in achieving learning outcomes. Learning analytics (LA) has matured in the education field and developed a strong connection with learning design. Learning analytics provides valuable insights to inform learning design...

Các tác giả

• Thi Hong Phuc Nguyen
• Ngoc Tram Nguyen Huynh
• Thi My Hang Vu

2024

Integrating Knowledge Graphs into English-Vietnamese Neural Machine Translation

The Transformer architecture, introduced since 2017, has demonstrated its power in machine translation tasks. However, significant challenges persist, notably in accurately understanding and converting meaning between languages. Out-of-vocabulary words, especially rare entities and terminological expressions in datasets, are the main cause of this inefficiency. We propose two methods integrating knowledge bases into the Transformer model to generate more accurate translations for entities and...

Các tác giả

• Nguyen Hong Buu Long
• Vo Thanh Phong

2024

Approaches for Extending Recommendation Models for Food Choices in Meals

In this paper, we propose food recommender systems based on users' historical food choices. Their advantage lies in providing personalized food suggestions for each user considering each meal. These systems are developed using two popular recommendation principles: neighbor-based and latent factor-based. In the neighbor-based model, the system aggregates the food choices of neighboring users to recommend food choices for the active user during the considered...

Các tác giả

• Tiet Gia Hong
• Vu Thị My Hang
• Nguyen Thi Hong Nhung
• Dao Khoa Nguyen

2024

Integrating Voice Activity Detection to Enhance Robustness of On-Device Speaker Verification

Mobile devices are integral to daily life, necessitating secure authentication methods like speaker verification for enhanced security and convenience. While deep neural networks have improved speaker verification performance, deploying these models on resource-constrained devices with low latency remains challenging. We address these issues by applying knowledge distillation to compress a speaker verification model for mobile deployment. Additionally, systems must handle long non-speech segments in audio...

Các tác giả

• Kiet Anh Hoang
• Khanh Duong
• Triet Nguyen Van Minh
• Tung Le & Huy Tien Nguyen

2024

Tackling Audio-Visual Condition Misalignment in Talking Head Generation model

The generation of talking landmarks from audio is pivotal for advancing talking head generation. This challenge poses a significant concern in landmark generation from audio and holds potential applications in various domains, including virtual assistants, education, and entertainment. However, existing audio-based methods exhibit limitations, such as inconsistencies in generated landmark frames and a lack of emotion features from the speech. In this...

Các tác giả

• Cao Xuan Nam
• Tran Minh Triet
• Trinh Quoc Huy

2024

Novel stochastic algorithms for privacy-preserving utility mining

High-utility itemset mining (HUIM) is a technique for extracting valuable insights from data. When dealing with sensitive information, HUIM can raise privacy concerns. As a result, privacy-preserving utility mining (PPUM) has become an important research direction. PPUM involves transforming quantitative transactional databases into sanitized versions that protect sensitive data while retaining useful patterns. Researchers have previously employed stochastic optimization methods to conceal sensitive patterns in databases through the addition or deletion of...

Các tác giả

• Duc Nguyen
• Bac Le

2024

StarSRGAN: Improving Real-World Blind Super-Resolution

The aim of blind super-resolution (SR) in computer vision is to improve the resolution of an image without prior knowledge of the degradation process that caused the image to be low-resolution. The State of the Art (SOTA) model Real-ESRGAN has advanced perceptual loss and produced visually compelling outcomes using more complex degradation models to simulate real-world degradations. However, there is still room to improve the superresolved quality of Real-ESRGAN by implementing recent...

Các tác giả

• Bui Tien Len
• Vo Dang Khoa

2024

A personalized learning path recommendation solution based on knowledge map mining

Recommendation systems (RS) are extensively used in various fields, especially in education, where intelligent e-learning platforms suggest personalized learning paths (PLP) tailored to learners and educational resources. Despite ongoing efforts to offer highly personalized recommendations, challenges like data sparsity and cold-start issues remain. Recently, the development of knowledge graph (KG)-based RS has attracted considerable attention. KGs can utilize semantic relationships between entities within a unified graph structure to address these...

Các tác giả

• Thu Minh Tran Nguyen
• Duong Thien Nguyen

2024

Lung Nodule Detection based on deep learning integrated with the Attention mechanism in CT Images

This research introduces a sophisticated technique for segmenting lung nodule in CT scans, employing the MHA-SEPA model, which is built upon the ResUNet++ framework. The MHA-SEPA architecture combines Multi-Head Attention mechanisms with SEBlock and Position Attention approaches to improve feature extraction by giving priority to important spatial and channel information. This method greatly enhances the precision of lung nodule segmentation, especially for tiny and difficult...

Các tác giả

• Thai Hoang Le
• Khai Dinh Lai
• and Thuy Thanh Nguyen

2024

Courses Recommender System for the IT field based on a context-aware knowledge model

This paper presents an approach for a knowledge-based recommender system that provides relevant courses based on learners’ profiles, requirements, and career needs. The framework integrates an automatic data collection process, ensuring that the knowledge base reflects the latest job market and course information. The recommendation method relies on a set of rules that combine various matching techniques, incorporating user requirements, skill and knowledge...

Các tác giả

• Pham Nguyen Cuong
• Le Nguyen Hoai Nam
• Pham Thi Xuan Hien

2024

Enhancing embeddings based on graph reasoning and time-aware attention networks on temporal knowledge graphs

Temporal Knowledge Graphs (TKGs) organize dynamic real-world facts, adding a time dimension to the multi-relational graph structure of Knowledge Graphs (KGs). We leverage the expressive power of graph convolutional networks (GCNs) for modeling TKGs, recognizing similarities with handling graph-structured data and utilizing complex geometry. Our approach emphasizes compositional interactions between relations and entities, integrating a diachronic mechanism to enhance representation with both graph structure and temporal...

Các tác giả

• Thanh Le
• Bac Le
• and Loc Tran

2024

An Approach for Estimating the Influence of Neighboring Users on the Target User in Recommender Systems

Recommendation systems play a crucial role in helping users navigate information overload, particularly in today's digital era. Their primary objective is to predict users' preferences for items. Latent factor-based recommendation systems achieve this by aligning users and items under latent factors. Previous studies mainly focused on devising effective objective functions for learning these latent factors. However, the accuracy of latent factors also depends on their initialization and the order of the collected data fed into the...

Các tác giả

• Le Nguyen Hoai Nam
• Ho Thi Hoang Vy
• Do Thi Thanh Ha
• Thi My Hang Vu
• Ho Le Thi Kim Nhung
• Cuong Pham-Nguyen
• Tiet Gia Hong

2024

Speaker verification and identification for secure and high utility Vietnamese virtual assistant

This paper presents our work in building a Vietnamese dataset for command and speaker recognition problems. We built a website that allows users of mobile devices or personal computers to be able to provide their voice samples easily. We collected more than fifteen thousand utterances of nineteen Vietnamese commands from more than two hundred volunteers. The commands are primarily used for applications on edge devices that interact with users via...

Các tác giả

• Minh-Nhut Ngo
• Long-Quoc Le

2024

A secure user authentication scheme for crypto-wallet in IoT environment

Presently, the prevalent environment is the Internet of things (IoT) since it can connect a large number of different gadgets in an increasingly wide range, and lead to the risk of information leakage. In addition, the field of crypto-currency is a hot topic with many sophisticated thefts, so the progressive utilization of crypto-wallets should protect transactions from eavesdropping. Clearly, IoT could be a promising and challenging environment for quick and secure cryptocurrency...

Các tác giả

• Truong Toan Thinh
• Tran Minh Triet
• Duong Anh Duc

2024

Integration of pre-trained Vision – Language model into Vietnamese Visual Question Answering

In recent decades, artificial intelligence has made significant progress in understanding and interacting with images. One of the impor- tant applications of this technology is Visual Question Answering (VQA), a research field that requires computers to understand and answer questions about images in a natural manner. Despite extensive research and development in VQA for English, there have been very few similar efforts made for other...

Các tác giả

• Le Thanh Tung
• Nguyen Tien Huy
• Nguyen Cong Phu

2024

CollaXRSearch: A Collaborative Virtual Reality System for Lifelog Retrieval

In lifelog data search, despite automatic supports for identifying relevant pieces of information, the processes of inputting queries and filtering information from the generated results still heavily rely on human searchers. With the rapid increase in volume of such data, these tasks could become both mentally and physically tedious for an individual to perform. In this paper, we present CollaXRSearch, a collaborative virtual reality (VR) information retrieval...

Các tác giả

• Khanh-Duy Le
• Duy-Nam Ly
• Dinh-Thuan Duong-Le
• Gia-Huy Vuong
• Van-Son Ho
• Van-Tu Ninh
• Minh-Triet Tran

2024

ArmorDroid: A Rule-Set Customizable Plugin for Secure Android Application Development

Although Android is a popular mobile operating system, its app ecosystem could be safer. The lack of awareness and concern for security issues in apps is one of the main reasons for this. Given the current situation, developers have yet to receive sufficient security knowledge. Therefore, we have researched and proposed a tool to support security coding. Based on the idea of...

Các tác giả

• Truong Phuoc Loc
• Tran Minh Triet
• Tran Anh Duy
• Le Cong Binh
• Nguyen Le Bao Thi

2024

Speaker authentication using deep neural network

This research deals with three challenges for speaker verification (SV): adaptivity, accuracy, and replay attack. We propose a framework consisting of three independent components: wakeword detector, one time password (OTP) block, and speaker identificator. With this architecture, we can customize each component without significant interference to the whole structure. Via these components, the final representation provides a meaningful information about the speaker to help the system verifies...

Các tác giả

• Chau Thanh Duc
• Do Duc Hao
• Van Khai Nguyen

2024

VIDES: Virtual Interior Design via Natural Language and Visual Guidance

Interior design is crucial in creating aesthetically pleasing and functional indoor spaces. However, developing and editing interior design concepts requires significant time and expertise. We propose Virtual Interior DESign (VIDES) system in response to this challenge. Leveraging cutting-edge technology in generative AI, our system can assist users in generating and editing indoor scene concepts quickly, given user text description and visual...

Các tác giả

• Le Trung Nghia
• Tran Minh Triet
• Hoang Nhu Vinh
• Le Minh Hien

2024

Vehicle Object Tracking Based on Fusing of Deep learning and Re-Identification

Object tracking is a popular problem for automatic surveillance systems as well as for the research community. The requirement of an object tracking problem is to predict the output including the object position at the current frame based on the input the position of the object at the previous frame. To present the comparison and experiment of some object tracking methods based on deep learning and suggestions for improvement between them in this...

Các tác giả

• Vo Hoai Viet
• Huynh Nhat Duy

2024

EAPC: Emotion and Audio Prior Control framework for the emotional and temporal Talking Face Generation

Generating realistic talking faces from audio input is a challenging task with broad applications in fields such as film production, gaming, and virtual reality. Previous approaches, employing a two-stage process of converting audio to landmarks and then landmarks to a face, have shown promise in creating vivid videos. However, they still face challenges in maintaining consistency due to misconnections between information from the previous audio frame and the current audio...

Các tác giả

• Cao Xuan Nam
• Tran Minh Triet
• Dang Hoai Thuong
• Trinh Quoc Huy
• Nguyen Do Quoc Anh
• Ho Van Son

2023

A syntax-aware deep-learning model for biomedical semantic role labelling

A deep learning model for biomedical semantic role labeling was build. Semantic role labeling is a useful task that enables the computer to comprehend the key facts expressed in each sentence, and is a necessary first step in the resolution of several other semantic-related tasks, such as event extraction, entity extraction, and Q-A systems... Semantic role labeling is a domain-dependent...

Các tác giả

• Tuan Nguyen Hoai Duc
• Nguyen Truong Son

2023

New Results on Erasure Combinatorial Batch Codes

We investigate in this work the problem of Erasure Combinatorial Batch Codes, in which n files are stored on m servers so that every set of n − r servers allows a client to retrieve at most k distinct files by downloading at most t files from each server. Previous studies have solved this problem for the special case of t =1 using Combinatorial Batch...

Các tác giả

• Le Phuc Lu
• Dau Son Hoang
• Ngo Dinh Hy
• Nguyen Dinh Thuc

2023

Exploring the Role of Monolingual Data in Cross-Attention Pre-training for Neural Machine Translation

Recent advancements in large pre-trained language models have revolutionized the field of natural language processing (NLP). Despite the impressive results achieved in various NLP tasks, their effectiveness in neural machine translation (NMT) remains limited. The main challenge lies in the mismatch between the pre-training objectives of the language model and the translation task, where the language modeling task focuses on reconstructing the language without considering its semantic interaction with other...

Các tác giả

• Dinh Dien
• Nguyen Hong Buu Long
• Pham Vinh Khang

2023

Exploring Graph-based Transformer Encoder for Low-Resource Neural Machine Translation

The Transformer is commonly used in Neural Machine Translation (NMT), but it faces issues with over-parameterization in low-resource settings. This means that simply increasing the model parameters significantly will not lead to improved performance. In this study, we propose a graph-based approach that slightly increases the parameters while significantly outperforming the scaled version of the Transformer. We accomplish this by utilizing Graph Neural Networks to encode Universal Conceptual Cognitive Annotation...

Các tác giả

• Nguyen Hong Buu Long
• Dinh Dien

2023

A new algorithm using integer programming relaxation for privacy-preserving in utility mining

High-utility itemset mining (HUIM) is an effective technique for discovering significant information in data. However, data containing sensitive and private information may cause privacy concerns. Therefore, privacy preserving utility mining (PPUM) has recently become a critical research area. PPUM is the process of transforming a quantitative transactional database into a sanitised one, thus ensuring that utility mining algorithms cannot discover sensitive...

Các tác giả

• Nguyen Ngoc Duc
• Le Hoai Bac

2023

A Robust Approach for Hybrid Personalized Recommender Systems

The personalization of services for users is one of the most crucial objectives of digital platforms. This objective is accomplished by integrating automated recommendation components into information systems. The increasing computational power and storage capacity available today have opened up opportunities to deploy a combination of diverse approaches to enhance the accuracy of the recommendation process. Compared to previous research, the distinguishing feature of this study is the introduction of an approach that combines not only computational aspects but also data...

Các tác giả

• Le Nguyen Hoai Nam
• Tiet Gia Hong

2023

Knowledge graph embedding by relational rotation and complex convolution for link prediction

Knowledge graphs are organized as triplets to represent facts from the real world and play an important role in various intelligent information systems. Because knowledge graphs are frequently constructed using manual or semi-automatic methods, they often miss connections between entities. Link prediction was created to solve this problem. Many recent state-of-the-art studies, such as those introducing the RotatE and RotatHS...

Các tác giả

• Le Ngoc Thanh
• Le Hoai Bac

2023

Toward Deep Transfer Learning for Realistic Activity Recognition in Videos

Today, videos have become popular on the internet and specified in social network and media platforms such as Youbue, Ticktok, and Vimeo. Video understanding has attracted much attention in the research community in recent years. Automatically recognizing human activity in wild videos is a trending research topic with a wide range of applications in advertising, smarthome, and surveillance camera systems. Deep convolutional neural networks have become a new de facto visual recognition...

Các tác giả

• Vo Hoai Viet

2023

GENA: A knowledge graph for nutrition and mental health

While a large number of knowledge graphs have previously been developed by automatically extracting and structuring knowledge from literature, there is currently no such knowledge graph that encodes relationships between food, biochemicals and mental illnesses, even though a large amount of knowledge about these relationships is available in the form of unstructured text in biomedical literature articles. To address this...

Các tác giả

• Phan Thi Phuong Uyen
• Nguyen Thi Hong Nhung
• Dang Diem Linh

2023

A Multi-Factor Approach to Measure User Preference Similarity in Neighbor-Based Recommender Systems

Neighbor-based Collaborative filtering is one of the commonly applied techniques in recommender systems. It is highly appreciated for its interpretability and ease of implementation. The effectiveness of neighbor-based collaborative filtering depends on the selection of a user preference similarity measure to identify neighbor users. In this paper, we propose a user preference similarity measure named Multi-Factor Preference Similarity (MFPS). The distinctive feature of our proposed method is its efficient combination of the four key factors in determining user preference similarity: rating...

Các tác giả

• Ho Thi Hoang Vy
• Tiet Gia Hong

2023

HybridMingler: Towards Mixed-Reality Support for Mingling at Hybrid Conferences

Mingling, the activity of ad-hoc, private, opportunistic conversations ahead of, during, or after breaks, is an important socializing activity for attendees at scheduled events, such as in-person conferences. The Covid-19 pandemic had a dramatic impact on the way conferences are organized, so that most of them now take place in a hybrid mode where people can either attend on-site or...

Các tác giả

• Le Khanh Duy
• Ly Duy Nam
• Le Quang Tri
• Nguyen Hoang Long

2022

Multilingual Communication System with Deaf Individuals Utilizing Natural and Visual Languages

According to the World Federation of the Deaf, more than two hundred sign languages exist. Therefore, it is challenging to understand deaf individuals, even proficient sign language users, resulting in a barrier between the deaf community and the rest of society. To bridge this language barrier, we propose a novel multilingual communication system, namely MUGCAT, to improve the communication efficiency of sign language...

Các tác giả

• Le Trung Nghia
• Huynh Tan Luc
• Chu Chi Bien
• Nguyen Ngoc Khoi Nguyen

2022

Meta-Learning and Personalization Layer in Federated Learning

Federated learning systems are confronted with two challenges: systemic and statistical. Non-IID data is acknowledged to be a primary component in causing statistical challenges. To address the federated learning system’s substantial performance loss on non-IID data, we offer the algorithm (which combines meta-learning methods and personalization layer approaches into a federated learning system). In terms of performance and personalization, has been shown in experiments to outperform typical federated learning...

Các tác giả

• Le Hoai Bac
• Nguyen Bao Long
• Cao Tat Cuong

2022

An approach to constructing a graph data repository for course recommendation based on IT career goals in the context of big data

Graph data is widely regarded as the next frontier in big data modeling for a variety of domains. Graph-based data models have been used in big data to organize messy or complicated data points based on their relationships. Graph analytic technologies for big data provide a framework for absorbing both structured and unstructured data from a variety of sources, allowing analysts to see the connections between entities in graphs and draw new...

Các tác giả

• Nguyen Tran Minh Thu
• Pham Minh Tu

2022

RVT-Transformer: Residual Attention in Answerability Prediction on Visual Question Answering for Blind People

Answerability Prediction on Visual Question Answering is an attractive and novel multi-modal task that can be regarded as a fundamental filter to eliminate the low-qualified samples in practical systems. Instead of focusing on the similarity between images and texts, the critical concern in this task is to accentuate the conflict in visual and textual information. However, the fusion function of the multi-modal system unwittingly decreases the original features of image and text that are essential in answerability...

Các tác giả

• Le Thanh Tung
• Nguyen Tien Huy

2022

Knowledge graph embedding by projection and rotation on hyperplanes for link prediction

Knowledge is increasingly completed due to connections formed in a knowledge graph, enabling a complete understanding of reality. Link prediction plays an important role in this process. Among the multiple methods that exist to tackle this problem, the geometry-based prediction method has attracted attention due to its intuitiveness and capacity to flexibly address various types of relations. We propose the rotation embedding of entities on separate relation-specific hyperplanes as an alternative to the translation...

Các tác giả

• Le Ngoc Thanh
• Le Hoai Bac

2022

MemInspect2: OS-Independent Memory Forensics for IoT Devices in Cybercrime Investigations

In the age of rapid development of the Internet of Things (IoT) world, more and more cybersecurity incidents have emerged in many IoT devices and systems. Therefore, the need for cybercrime investigation, especially for IoT devices, has become more imperative than ever. Memory forensics, the approach that inspects the memory dump to understand the current state or behavior of the attacked...

Các tác giả

• Tran Anh Duy
• Nguyen Quoc Trung
• Nguyen Anh Minh

2022

Research and development of face recognition system for attendance and monitoring of students in the classroom and exams

One of the primary activities that lectures usually do is to take a roll call. This activity not only helps lecturers determine the participation of students but also detect strangers in the classroom. When the number of students increases, lectures take more time to monitor and check students’ attendance. We propose a student monitoring system based on facial recognition approaches to tackle that...

Các tác giả

• Pham Trong Nghia
• Le Ngoc Thanh

2022

A semi-automatic approach for pasbio corpus data augmentation

A semi-automatic solution to build a biomedical semantic role corpus named PASBio+ was proposed. The corpus was annotated with a predicate argument structure, the important information that revealed the main content of a sentence. Because more than 86% of the arguments in the biomedical domain significantly differed from those in the general domain, this proposed corpus was labeled on top of 317 labeled sentences from...

Các tác giả

• Tuan Nguyen Hoai Duc

2022

Medical Prescription Recognition Using Heuristic Clustering and Similarity Search

The necessity to convert printed documents to facilitate the storage and retrieval of information is growing, particularly in the medi- cal and healthcare industries. In our last work, we presented a method to extract prescriptions from images using CRAFT and TESSERACT so that patients could quickly save and check up on their pharmaceutical use information. However, the slow processing speed and the limited number of medication names lead to it being...

Các tác giả

• Nguyen Ngoc Thao
• Le Ngoc Thanh

2022

M.A.R – Micro Automatic Robot

Robotics is one of the important subjects in automation and modernizing our country. The automatic robot has a variety of sizes and shapes. This helps people in any context, like discovering small or dangerous areas in collapsed houses, and caves or check the cash, leak in oil or water pipelines. To create a passion for science and application for the...

Các tác giả

• Cao Xuan Nam
• Nguyen Do Quoc Anh

2022

Binary Classification for Lung Nodule Based on Channel Attention Mechanism

In order to effectively handle the problem of tumor detection on the LUNA16 dataset, we present a new methodology for data augmentation to address the issue of imbalance between the number of positive and negative candidates in this study. Furthermore, a new deep learning model - ASS (a model that combines Convnet sub-attention with Softmax loss) is also proposed and evaluated on patches with different sizes of the...

Các tác giả

• Le Hoang Thai
• Lai Dinh Khai

2022

Group Testing with Blocks of Positives and Inhibitors

The main goal of group testing is to identify a small number of specific items among a large population of items. In this paper, we consider specific items as positives and inhibitors and non-specific items as negatives. In particular, we consider a novel model called group testing with blocks of positives and inhibitors. A test on a subset of items is positive if the subset contains at least one positive and does not contain any...

Các tác giả

• Nguyen Dinh Thuc
• Bui Van Thach

2022

ITCareerBot: A Personalized Career Counselling Chatbot

Nowadays, career counselling, which is a service designed to help people finding the right professional learning is emerging. In the information technology (IT) domain, this service is facing the challenge of changing very quickly in business environments, technologies and tools. Consequently, IT students and professionals often need additional knowledge and skills to fulfill market requirements and to target their professional goal in order to increase their opportunity for...

Các tác giả

• Pham Nguyen Cuong
• Le Nguyen Hoai Nam
• Nguyen Duy Cuong
• Dinh Nguyen Hanh Dung

2021

Object-less Vision-language Model on Visual Question Classification for Blind People

Despite the long-standing appearance of question types in the Visual Question Answering dataset, Visual Question Classification does not received enough public interest in research. Different from general text clas-sification, a visual question requires an understanding of visual and textual features simultaneously. Together with the enthusiasm and novelty of Visual Question Classification, the most important and practical goal we concentrate on is to deal with the weakness of Object Detection on object-less...

Các tác giả

• Nguyen Tien Huy
• Bui Huy Thong

2021

Using Bert embedding to improve memory-based collaborative recommender systems

The performance of memory-based collaborative filtering recommender systems will be severely affected when the users' item preference data is sparse. In this paper, we focus on solving this issue. Our idea is to use Bert Embedding to learn a new feature set, which is denser and more semantic, for representing users and items. In these new features, memory based collaborative filtering recommender systems work more ...

Các tác giả

• Ho Thi Hoang Vy
• Le Nguyen Hoai Nam

Đối tác doanh nghiệp