Emotion recognition is becoming increasingly important for accurately understanding and responding to user emotions, driven by the rapid proliferation of non-face-to-face environments and advancements in conversational AI technologies. Existing studies on multimodal emotion recognition, which utilize text and speech, have aimed to improve performance by integrating the information from both modalities. However, these approaches have faced limitations such as restricted information exchange and the omission of critical cues. To address these challenges, this study proposes a Hybrid Multimodal Transformer, which combines Intermediate Layer Fusion and Last Fusion. Text features are extracted using KoELECTRA, while speech features are extracted using HuBERT. These features are processed through a transformer encoder, and Dual Cross Modal Attention is applied to enhance the interaction between text and speech. Finally, the predicted results from each modality are aggregated using an average ensemble method to recognize the final emotion. The experimental results indicate that the proposed model achieves superior emotion recognition performance compared to existing models, demonstrating significant progress in improving both the accuracy and reliability of emotion recognition. In the future, incorporating additional modalities, such as facial expression recognition, is expected to further strengthen multimodal emotion recognition capabilities and open new possibilities for application across diverse fields.

With the rising number of Internet users, there has been a rapid increase in cyberbullying. Among the types of cyberbullying, verbal abuse is emerging as the most serious problem, for preventing which profanity is being identified and blocked. However, users employ words cleverly to avoid blocking. With the existing profanity discrimination methods, deliberate typos and profanity using special characters can be discriminated with high accuracy. However, as they cannot grasp the meaning of the words and the flow of sentences, standard words such as “Sibaljeom (starting point, a Korean word that sounds similar to a swear word)” and “Saekkibalgalag (little toe, a Korean word that sounds similar to another swear word)” are less accurately discriminated. Therefore, in order to solve this problem, this study proposes a method of discriminating profanity using a deep learning model that can grasp the meaning and context of words after separating Hangul into the onset, nucleus, and coda.

The megatrends and Industry 4.0 in ICT (Information Communication & Technology) are concentrated in IoT (Internet of Things), BigData, CPS (Cyber Physical System), and AI (Artificial Intelligence). These megatrends do not operate independently, and mass storage technology is essential as large computing technology is needed in the background to support them. In order to evaluate the performance of high-capacity storage based on open source Ceph, we carry out the network performance test of Abyss storage with domestic and overseas sites using KOREN (Korea Advanced Research Network). And storage media and network bonding are tested to evaluate the performance of the storage itself. Additionally, the security test is demonstrated by Cuckoo sandbox and Yara malware detection among Abyss storage cluster and oversea sites. Lastly, we have proposed the draft design of Data Lake framework in order to solve garbage dump problem.

Recently, multimodal approaches that combine various modalities have been attracting attention to recognizing emotions more accurately. Although multimodal fusion delivers strong performance, it is computationally intensive and difficult to handle in real time. In addition, there is a fundamental lack of large-scale emotional datasets for learning. In particular, Korean emotional datasets have fewer resources available than English-speaking datasets, thereby limiting the generalization capability of emotion recognition models. In this study, we propose a more lightweight modality fusion method, MMER-LMF, to overcome the lack of Korean emotional datasets and improve emotional recognition performance while reducing model training complexity. To this end, we suggest three algorithms that fuse emotion scores based on the reliability of each model, including text emotion scores extracted using a pre-trained large-scale language model and video emotion scores extracted based on a 3D CNN model. Each algorithm showed similar classification performance except for slight differences in disgust emotion performance with confidence-based weight adjustment, correlation coefficient utilization, and the Dempster–Shafer Theory-based combination method. The accuracy was 80% and the recall was 79%, which is higher than 58% using text modality and 72% using video modality. This is a superior result in terms of learning complexity and performance compared to previous studies using Korean datasets.