The COVID-19 pandemic has led to a surge in online education, yielding increased attention on knowledge tracing (KT). KT involves predicting a student’s mastery of knowledge based on the student’s ...learning activity and plays a crucial role for recommending appropriate educational content at the individual level. While several deep learning-based KT models have been proposed, most consider only the interaction sequence of the target student, ignoring the interactions of other students. Furthermore, in practice, all model graphs should be incremented as more student interaction data is obtained, implying that they should be rebuilt to remain up-to-date, which is time-consuming. Moreover, existing graph-based KT models typically rely on predefined knowledge graphs for graph-based learning. Constructing a knowledge graph is labor-intensive, requiring a higher understanding of the domain knowledge. To address these problems, we propose a new inductive graph-based KT (IGKT) framework, which fully utilizes a subgraph extraction approach from a student-exercise bipartite graph, promising scalability and efficiency for practical applicability. The proposed framework consists of three key components: subsequence extraction, subgraph extraction and exericse information addition, and graph-based correctness prediction. For the third component, we test three variants to determine the effectiveness of integrating the timestamp and graph information, thereby creating three models for comparison: IGKT-Basic, IGKT with timestamps (IGKT-TS), and IGKT with graph attention layers that incorporate timestamps as edge features (IGKT-GAT). Extensive experiments conducted on multiple real-world datasets show that the models based on the proposed framework are superior over state-of-the-art models, and IGKT-GAT is the most effective implementation of the proposed framework. We further validate the effectiveness of the IGKT models through an ablation study and multiple in-depth analyses.
Graphical abstract
Sensor data is structured and generally lacks of meaning by itself, but life-logging data (time, location, etc.) out of sensor data can be utilized to create lots of meaningful information combined ...with social data from social networks like Facebook and Twitter. There have been many platforms to produce meaningful information and support human behavior and context-awareness through integrating diverse mobile, social, and sensing input streams. The problem is that these platforms do not guarantee the performance in terms of the processing time and even let the accuracy of output data be addressed by new studies in each area where the platform is applied. Thus, this study proposes an improved platform which builds a knowledge base for context awareness by applying distributed and parallel computing approach considering the characteristics of sensor data that is collected and processed in real-time, and compares the proposed platform with existing platforms in terms of performance. The experiment shows the proposed platform is an advanced platform in terms of processing time. We reduce the processing time by 40% compared with existing platform. The proposed platform also guarantees the accuracy compared with existing platform.
Objectives
To evaluate the safety and efficacy of expired lyophilized snake antivenom of Thai origin during a medical emergency in 2020/2021 in Lao People's Democratic Republic.
Methods
Observational ...case series of patients with potentially life‐threatening envenoming who consented to the administration of expired antivenom between August 2020 and May 2022.
Results
A total of 31 patients received the expired antivenom. Malayan pit vipers (Calloselasma rhodostoma) were responsible for 26 (84%) cases and green pit vipers (Trimeresurus species) for two cases (6%). In three patients (10%) the responsible snake could not be identified. Of these, two presented with signs of neurotoxicity and one with coagulopathy. A total of 124 vials of expired antivenom were administered. Fifty‐nine vials had expired 2–18 months earlier, 56 vials 19–36 months and nine vials 37–60 months before. Adverse effects of variable severity were observed in seven (23%) patients. All 31 patients fully recovered from systemic envenoming.
Conclusions
Under closely controlled conditions and monitoring the use of expired snake antivenom proved to be effective and safe. Discarding this precious medication is an unnecessary waste, and it could be a valuable resource in ameliorating the current shortage of antivenom. Emergency use authorization granted by health authorities and preclinical testing of expired antivenoms could provide the support and legal basis for such an approach.
Inductive Graph-based Knowledge Tracing Han, Donghee; Kim, Daehee; Han, Keejun ...
2023 IEEE International Conference on Big Data and Smart Computing (BigComp),
2023-Feb.
Conference Proceeding
The rise of virtual education and increase in distance, partly owing to the spread of COVID-19 pandemic, has made it more difficult for teachers to determine each student's learning status. In this ...situation, knowledge tracing (KT), which tracks a student's mastery of specific knowledge concepts, is receiving increasing attention. KT utilizes a sequence of studentexercise interactive activities to predict the mastery of concepts corresponding to a target problem, recommending appropriate learning resources to students and optimizing learning sequences for adaptive learning. With the development of deep learning, various studies have been proposed, such as sequential models using recurrent neural networks, attention models influenced by transformers, and graph-based models that depict the relationships between knowledge concepts. However, they all have common limitations in that they cannot utilize the learning activities of students other than the target student and can only use a limited form of exercise information. In this study, we have applied the concept of rating prediction to the studentexercise knowledge tracing problem and solved the limitations of the existing models. Our proposed Inductive Graph-based Knowledge Tracing (IGKT) designed to integrate structural information and various unrestricted types of additional information into the model through subgraph sampling, has been found superior over the existing models across two different datasets in predicting student performances.
Engineers create engineering documents with their own terminologies, and want to search existing engineering documents quickly and accurately during a product development process. Keyword-based ...search methods have been widely used due to their ease of use, but their search accuracy has been often problematic because of the semantic ambiguity of terminologies in engineering documents and queries. The semantic ambiguity can be alleviated by using a domain ontology. Also, if queries are expanded to incorporate the engineer’s personalized information needs, the accuracy of the search result would be improved. Therefore, we propose a framework to search engineering documents with less semantic ambiguity and more focus on each engineer’s personalized information needs. The framework includes four processes: (1) developing a domain ontology, (2) indexing engineering documents, (3) learning user profiles, and (4) performing personalized query expansion and retrieval. A domain ontology is developed based on product structure information and engineering documents. Using the domain ontology, terminologies in documents are disambiguated and indexed. Also, a user profile is generated from the domain ontology. By user profile learning, user’s interests are captured from the relevant documents. During a personalized query expansion process, the learned user profile is used to reflect user’s interests. Simultaneously, user’s searching intent, which is implicitly inferred from the user’s task context, is also considered. To retrieve relevant documents, an expanded query in which both user’s interests and intents are reflected is then matched against the document collection. The experimental results show that the proposed approach can substantially outperform both the keyword-based approach and the existing query expansion method in retrieving engineering documents. Reflecting a user’s information needs precisely has been identified to be the most important factor underlying this notable improvement.
Plants have been the most important natural resources for traditional medicine and for the modern pharmaceutical industry. They have been in demand in regards to finding alternative medicinal herbs ...with similar efficacy. Due to the very low probability of discovering useful compounds by random screening, researchers have advocated for using targeted selection approaches. Furthermore, because drug repositioning can speed up the process of drug development, an integrated technique that exploits chemical, genetic, and disease information has been recently developed. Building upon these findings, in this paper, we propose a novel framework for the targeted selection of herbs with similar efficacy by exploiting drug repositioning technique and curated modern scientific biomedical knowledge, with the goal of improving the possibility of inferring the traditional empirical ethno-pharmacological knowledge.
To rank candidate herbs on the basis of similarities against target herb, we proposed and evaluated a framework that is comprised of the following four layers: links, extract, similarity, and model. In the framework, multiple databases are linked to build an herb-compound-protein-disease network which was composed of one tripartite network and two bipartite networks allowing comprehensive and detailed information to be extracted. Further, various similarity scores between herbs are calculated, and then prediction models are trained and tested on the basis of theses similarity features.
The proposed framework has been found to be feasible in terms of link loss. Out of the 50 similarities, the best one enhanced the performance of ranking herbs with similar efficacy by about 120–320% compared with our previous study. Also, the prediction model showed improved performance by about 180–480%. While building the prediction model, we identified the compound information as being the most important knowledge source and structural similarity as the most useful measure.
In the proposed framework, we took the knowledge of herbal medicine, chemistry, biology, and medicine into consideration to rank herbs with similar efficacy in candidates. The experimental results demonstrated that the performances of framework outperformed the baselines and identified the important knowledge source and useful similarity measure.
Display omitted
The neurotoxicity of immunosuppressive agents and diabetes mellitus are known risk factors of neurological complications in kidney transplant recipients. The aim of the present study was to ...investigate the influence of tacrolimus on brain-derived neurotrophic factor (BDNF), the critical protein for maintenance of neuronal functions, in the hippocampus in a diabetic condition. A diabetic rat model was established by a single streptozotocin injection (60 mg/kg). Control and diabetic rats then received daily tacrolimus (1.5 mg/kg per day) injections for 6 weeks. BDNF expression in the hippocampus was examined in the dentate gyrus (DG) and CA3 region using immunohistochemistry. There was a significant decrease of BDNF expression in the DG and CA3 region in tacrolimus-treated and diabetic rats compared with that of the control group injected with vehicle only. However, there was no difference in BDNF expression between the two experimental groups. Tacrolimus treatment in diabetic rats further decreased the BDNF expression level in the DG and CA3 region. Interestingly, mossy fiber sprouting, demonstrated by prominent punctate immunolabeling of BDNF with synaptoporin, was observed in the diabetic group treated with tacrolimus, which localized at the stratum oriens of the CA3 region. These data suggest that tacrolimus treatment or a diabetic condition decreases BDNF expression in the hippocampus, and that tacrolimus treatment in the diabetic condition further injures the CA3 region of the hippocampus. In addition to BDNF expression, decreased locomotor activity and evident depressive behavior were observed in tacrolimus-treated diabetic rats. Moreover, there were significant decreases of the mRNA levels of γ-aminobutyric acid and serotonin receptors in the diabetic hippocampus with tacrolimus treatment. This finding suggests that tacrolimus treatment may cause further psychiatric and neurological complications for patients with diabetes, and should thus be used with caution.
Semantic annotation approaches link entities from a knowledge base to mentions of entities in text to provide additional content-related information. Recently increasing use of resources from the ...Linked Open Data (LOD) Cloud has been made to annotate text documents thanks to the network of machine-understandable, interlinked data. While existing approaches to semantic annotation in the LOD context have been proven to be well performing with the English language, many other languages in general and the Korean language in particular are still underrepresented. We investigate the applicability of existing semantic annotation approaches to the Korean language by adapting two popular approaches in the semantic annotation field and evaluating those approaches on an English-Korean bilingual sense-tagged corpus. Further, general challenges in internationalization of annotation approaches are summarized.