Search

Entity alignment (EA) aims to discover the equivalent entities in different knowledge graphs (KGs), which plays an important role in knowledge engineering. Recently, EA with dangling entities has been proposed as a more realistic setting, which assumes that not all entities have corresponding equivalent entities. In this paper, we focus on this setting. Some work has explored this problem by leveraging translation API, pre-trained word embeddings, and other off-the-shelf tools. However, these approaches over-rely on the side information (e.g., entity names) and fail to work when the side information is absent. On the contrary, they still insufficiently exploit the most fundamental graph structure information in KG. To improve the exploitation of the structural information, we propose a novel entity alignment framework called Structure-aware Wasserstein Graph Contrastive Learning (SWGCL), which is refined on three dimensions: (i) Model. We propose a novel Gated Graph Attention Network to capture local and global graph structure attention. (ii) Training. Two learning objectives: contrastive learning and optimal transport learning, are designed to obtain distinguishable entity representations. (iii) Inference. In the inference phase, a PageRank-based method HOSS (Higher-Order Structural Similarity) is proposed to calculate higher-order graph structural similarity. Extensive experiments on two dangling benchmarks demonstrate that our SWGCL outperforms the current state-of-the-art methods with pure structural information in both traditional (relaxed) and dangling (consolidated) settings. [ABSTRACT FROM AUTHOR]

Recently, prompt-based methods have achieved a promising performance in many natural language processing benchmarks. Despite success in sentence-level classification tasks, prompt-based methods work poorly in token-level tasks, such as named entity recognition (NER), due to the sophisticated design of entity-related templates. Note that the nature of prompt tuning makes full use of the parameters of the mask language model (MLM) head, while previous methods solely utilized the last hidden layer of language models (LMs) and the power of the MLM head is overlooked. In this work, we discovered the characteristics of semantic feature changes in samples after being processed using MLMs. Based on this characteristic, we designed a prompt-tuning variant for NER tasks. We let the pre-trained model predict the label words derived from the training dataset at each position and fed the generated logits (non-normalized probability) to the CRF layer. We evaluated our method on three popular datasets, and the experiments showed that our proposed method outperforms the state-of-the-art model in all three Chinese datasets. [ABSTRACT FROM AUTHOR]

Purpose: Breast cancer is one of the most common malignant cancers in women, seriously endangering the physical and mental health of patients. In this study, we developed an app for breast cancer patients undergoing radiotherapy or chemotherapy with a focus on exercise interventions, supplemented by nutritional and psychological interventions, to verify the applicability of the app for these patients and its impact on their quality of life, sleep, and psychological state. We also investigated the patients' experience and perceptions of the app. Methods: A total of 17 participants, aged 42-58years, were recruited for this study using a mixed-methods design, including quantitative group pre-and posttest scores and qualitative interview results. The participants used the app for 8-18weeks depending on their radiotherapy or chemotherapy cycle. During the radiotherapy or chemotherapy period, the participants used the "Yun Dong Ru Kang" exercise rehabilitation app to perform aerobic exercises twice a week, as well as rehabilitation exercises appropriate to their radiotherapy or chemotherapy stage, and used the app on their own the rest of the time. The primary results included their scores on the PSSUQ overall assessment usability questionnaire, the users' use of the app, and the results of the interviews; the secondary indicators were quality of life, sleep status, and anxiety and depression status. Results: An overall score of 6.2 (out of 7 points) on the PSSUQ questionnaire indicates the high usability; the average use time per subject per week was 97.69±11.82min, which exceeds the minimum use time, but the average use time tended to decrease as the use time was postponed. Promoted articles on nutritional diets received the most hits. The results of the interviews were consistent with the questionnaire scores, with the majority of participants believing that the means of exercise should be enriched and the interface optimized, while the reduction in the length of use was related to the participants' own state of learning about calisthenics. In the results of the Breast Cancer-Specific Scale FACT-B, there was a significant increase (p<0.05) in the Emotional Status dimension score and a significant decrease (p<0.05) on the Additional Concerns dimension score. In the results of the Pittsburgh Sleep Quality Inventory PSQI, there was a non-significant improvement in all items except for a significant increase (p<0.05) for the Hypnotic Medication item. In the Hospital Anxiety and Depression Scale (HADS), there was no significant improvement in any of the anxiety and depression factors. Conclusions: The Yun Dong Ru Kang app has certain applicability, and the use of the exercise rehabilitation app may effectively reduce the negative impact of chemotherapy side effects on the quality of life, sleep and depression of breast cancer patients in the chemotherapy or radiotherapy phase. Before it is put into use in the future, the app should be enriched with exercise tools, the interface should be optimized, and articles on nutrition and diet should be promoted. [ABSTRACT FROM AUTHOR]