Dukjo Kong

Prof. Dukjo Kong

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 May 2024 in IEEE Access
Verified
Journal Article
Open access
GLGeon Woo Lee
Geon Woo Lee
Hong Kook KimHong Kook Kim
Hong Kook Kim
DKDukjo Kong
Dukjo Kong
This paper addresses the training issues associated with neural network-based automatic speech recognition (ASR) under noise conditions. In particular, conventional joint training approaches for a pipeline comprising speech enhancement (SE) and end-to-end ASR model surfer from a conflicting problem and a frame mismatched alignment problem because of different goals and different frame structures for ASR and SE. To mitigate such problems, a knowledge distillation (KD)-based training approach is proposed by interpreting the ASR and SE models in the pipeline as teacher and student models, respectively. In the proposed KD-based training approach, the ASR model is first trained using a training dataset, and then, acoustic tokens are generated via K-means clustering using the latent vectors of the ASR encoder. Thereafter, KD-based training of the SE model is performed using the generated acoustic tokens. The performance of the SE and ASR models is evaluated on two different databases, noisy LibriSpeech and CHiME-4, which correspond to simulated and real-world noise conditions, respectively. The experimental results show that the proposed KD-based training approach yields a lower character error rate (CER) and word error rate (WER) on the two datasets than conventional joint training approaches, including multi-condition training. The results also show that the speech quality scores of the SE model trained using the proposed training approach are higher than those of SE models trained using conventional training approaches. Moreover, the noise reduction scores of the proposed training approach are higher than those of conventional joint training approaches but slightly lower than those of the standalone-SE training approach. Finally, an ablation study is conducted to examine the contribution of different combinations of loss functions in the proposed training approach to SE and ASR performance. The results show that the combination of all loss functions yields the lowest CER and WER and that tokenizer loss contributes more to SE and ASR performance improvement than ASR encoder loss.

5 June 2020 in ACS Applied Nano Materials
Verified
Journal Article
Closed access
YYYoung Jin Yoo
Young Jin Yoo
JKJoo Hwan Ko
Joo Hwan Ko
WKWon-Guen Kim
Won-Guen Kim
YKYeong Jae Kim
Yeong Jae Kim
DKDukjo Kong
Dukjo Kong
Young Min SongYoung Min Song
Young Min Song

1 March 2024 in Nanophotonics
Verified
Journal Article
Open access
PDF file
SCSehui Chang
Sehui Chang
DKDukjo Kong
Dukjo Kong
Young Min SongYoung Min Song
Young Min Song
Artificial vision systems pervade our daily lives as a foremost sensing apparatus in various digital technologies, from smartphones to autonomous cars and robotics. The broad range of applications for conventional vision systems requires facile adaptation under extreme and dynamic visual environments. However, these current needs have complicated individual visual components for high‐quality image acquisition and processing, which indeed leads to a decline in efficiency in the overall system. Here, we review recent advancements in visual components for high‐performance visual processing based on strategies of biological eyes that execute diverse imaging functionalities and sophisticated visual processes with simple and concise ocular structures. This review first covers the structures and functions of biological eyes (i.e., single‐lens eyes and compound eyes), which contain micro‐optic components and nanophotonic structures. After that, we focus on their inspirations in imaging optics/photonics, light‐trapping and filtering components, and retinomorphic devices. We discuss the remaining challenges and notable biological structures waiting to be implemented.

19 June 2024 in International Journal of Human–Computer Interaction
Verified
Journal Article
Open access
Sanghyub LeeSanghyub Lee
Sanghyub Lee
WKWoojin Kang
Woojin Kang
JHJin-Hyuk Hong
Jin-Hyuk Hong
DKDukjo Kong
Dukjo Kong

28 January 2026 in Analytical Methods
Verified
Journal Article
Open access
PDF file
SSSungho Shin
Sungho Shin
JPJohn Park
John Park
DKDukjo Kong
Dukjo Kong
A combined LIBS-Raman platform enables simultaneous molecular and elemental analysis of microplastics. Prototype testing verified accurate classification and quantitative analysis, demonstrating feasibility result for on-site monitoring.

1 April 2026 in Journal of Medical Internet Research
Verified
Journal Article
Open access
JLJihye Lee
Jihye Lee
JPJuyoung Park
Juyoung Park
YKYuna Kim
Yuna Kim
DKDukjo Kong
Dukjo Kong
Background: Preventive campaigns for older adults must decide how to allocate limited resources across media channels. However, these channel allocation and budget decisions rarely use explicit criteria for distributional equity or structured strategic planning tools. Consequently, health systems may optimize average uptake while leaving large gaps across socioeconomic groups and media use profiles. Objective: This study aimed to develop and apply a data-driven agent-based model as a strategic planning tool for preventive campaigns targeting older adults, comparing channel allocation, personalization, and loss framing options under explicit budget and equity guardrails. Methods: We built an agent-based model calibrated to national survey data from South Korea on influenza vaccination and routine health screening among older adults (vaccination, N=2405; screening, N=2400). Fifteen prespecified campaign scenarios varied channel allocation across television, digital, and print media; budget intensity; 2 equity-focused personalization strategies; and graded loss framing. Primary outcomes were final adoption and time to adoption. Equity outcomes included the minimum class-level adoption and 90‐10 gap across latent classes. Each scenario was simulated over 12 monthly steps with 100 Monte Carlo replications. We conducted sensitivity analyses varying link functions and key social reinforcement parameters. Results: Personalization improved uptake and equity relative to the integrated baseline. In the vaccination model (N=2405), adoption increased from 91.2% (n=2193) to 93.3% (n=2244) and 94.6% (n=2275). Minimum class-level adoption increased from 86.8% to 90.3% and 90.9%. The 90‐10 gap narrowed from 5.7 to 4.5 and 4.7 percentage points. In the screening model (N=2400), adoption increased from 83.8% (n=2011) to 88.2% (n=2117) and 89.5% (n=2148). Minimum class-level adoption increased from 77.6% to 83.2% and 85.3%. The 90‐10 gap narrowed from 9.2 to 7.4 and 6.2 percentage points. Television-only strategies achieved high adoption but had less favorable equity profiles than personalization. High-budget strategies achieved high adoption but required higher total exposure. Stronger loss framing produced small, monotonic gains in adoption and shortened the time to adoption without worsening equity in the tested range. Scenario rankings were stable in sensitivity analyses. Conclusions: This agent-based modeling study illustrates how ex ante planning can improve preventive campaign design by comparing channel allocation and personalization options under explicit equity and budget criteria. For campaigns targeting older adults, equity-focused reweighting and class-tailored television-digital portfolios improved or preserved mean adoption while strengthening distributional equity under fixed budgets. In contrast, undifferentiated channel diversification without personalization offered a less favorable efficiency-equity trade-off. These findings support integrating explicit equity guardrails into early-stage channel allocation and prioritizing targeted personalization over simple channel diversification. Future work should validate these patterns in other populations and health systems and link simulated diffusion trajectories with observed exposure and engagement in real-world campaigns. It should also extend guardrail-based planning tools to organizational settings and multiyear decision contexts.