Code-switching is a phenomenon of alternating use of two or more languages within or between utterances in communication that often occurs in multilingual communities. Recently, code-switching natural language processing and automatic speech recognition (ASR) have attracted numerous studies. However, a major obstacle affecting the results of these studies is the lack of transcribed data. In this letter, we propose a novel semi-supervised learning (SSL) approach to deal with this problem, namely Mutual Learning-Based Semi-Supervised Method (MLSS). The MLSS method involves the utilization of two networks for interleaved fine-tuning on a combination of transcribed dataset and pseudo-labeled data generated from another network. This iterative fine-tuning process repeats until all unlabeled data are selected for training or reaches a certain number of iterations. By incorporating mutual learning between the two networks, our approach effectively leverages the knowledge acquired from previous iterations during the training stage and combines the knowledge from both networks during the decoding process, resulting in a more robust and effective approach. To evaluate the effectiveness of our proposed method, we conduct experiments on the SEAME Mandarin-English code-switching corpus. The experimental results clearly illustrate that our approach outperforms other state-of-the-art methods, as evidenced by achieving a Mixed Error Rate (MER) of 15.6% /21.1% on test$_{man}$/test$_{sge}$sets.

Abstract The role of medical test results in the diagnosis and treatment of a patient’s disease cannot be denied. Doctors and medical staff rely on these results to develop a treatment plan that meets the requirements of the patient’s health status (i.e., physical condition) and disease type. In developing countries (i.e., Vietnam), we note that test results are recorded in paper versions and stored by patients. Thus, several solutions have been introduced for electronic medical records to compensate for medical test results management. However, in Vietnam, these approaches face many obstacles such as centralized processing (e.g., storage, analysis); non-transparency issues; scalability; availability; and so on. In this paper, we exploit the benefits of blockchain, smart contracts, and NFT technologies to solve the above disadvantages. Therefore, our work contributes to five aspects. (a) Collecting procedures for handling and storing test results of patients at hospitals in Ho Chi Minh City and Mekong Delta (i.e., Can Tho city) (b) Proposing a mechanism for sharing test results based on blockchain technology, smart contract, and NFT applied; (c) Presenting an NFT tool-based certification generation model; (d) implementing the proposed model based on smart contracts (i.e., proof-of-concept); and (e) deploying proof-of-concept on four EVM- and NFT-supported platforms to find the most suitable one.

Abstract The end of the covid-19 epidemic has revealed many weaknesses in the health system and the medical waste treatment process. In particular, the ineffective treatment of medical waste has also contributed to the explosion in the number of infections in some countries (i.e., India, Brazil, and Vietnam). Several studies have found that even developed countries (i.e., with better infrastructure and health services than the world average) have to face emergencies during this time. epidemic. Therefore, the amount of medical waste dumped into the environment is extremely terrible. The waste generated suddenly during this period includes protective gear, masks, and vaccines that burden the waste treatment process. There are several approaches to exploiting Blockchain technologies to solve the problem of direct contact between the stages: medical staff - transportation staff - waste disposal staff to minimize unintended spread. However, to thoroughly solve the current waste classification and treatment processes, a more reward/punishment solution is needed. Specifically, we propose a model to assess the compliance/violation level of waste sorting and treatment in medical centers and isolation areas based on current popular technologies: blockchain, smart contracts, and NFTs.

Transfer learning is a common method to improve the performance of the model on a target task via pre-training the model on pretext tasks. Different from the methods using monolingual corpora for pre-training, in this study, we propose a Cyclic Transfer Learning method (CTL) that utilizes both code-switching (CS) and monolingual speech resources as the pretext tasks. Moreover, the model in our approach is always alternately learned among these tasks. This helps our model can improve its performance via maintaining CS features during transferring knowledge. The experiment results on the standard SEAME Mandarin-English CS corpus have shown that our proposed CTL approach achieves the best performance with Mixed Error Rate (MER) of 16.3% on test$_{man}$, 24.1% on test$_{sge}$. In comparison to the baseline model that was pre-trained with monolingual data, our CTL method achieves 11.4% and 8.7% relative MER reduction on the test$_{man}$and test$_{sge}$sets, respectively. Besides, the CTL approach also outperforms compared to other state-of-the-art methods. The source code of the CTL method can be found athttps://github.com/caohongnga/CTL-CSSR.

Summary The spread of brown plant hoppers (BPHs) can be affected by pesticides, migration/immigration, and natural enemies; in particular, using pesticides is the common method of farmers in fact. The article introduces a new approach to simulate the spread of BPH using cellular automata. The proposed model uses a geographic information system map as a simulation space to make results more visual and realistic. This model can give the accurate predictions about the spread of BPHs, thereby helping farmers make the appropriate decisions to prevent BPHs.