
Application usage prediction is a crucial technique for optimizing smartphone systems.This study proposes a semantic-en-hanced prediction method based on large language models to improve the understanding of user intent and generalization to out-of-distribution applications.The method used a large language model to automatically generate functional descriptions of applications.These descriptions were encoded into high-dimensional semantic vectors through an embedding model and used as semantic repre-sentations of historical usage sequences.These representations were then fed into the prediction model.To better capture temporal dynamics and semantic dependencies in user behavior,a prediction model was designed by integrating locally activated attention mechanisms with a Mamba-2 module.Experimental results showed that on a real-world application usage dataset,the proposed method increased the Top-3 accuracy from 76.41%to 77.00%on in-distribution data and from 66.06%to 71.57%on out-of-distribu-tion data,significantly improving the model's generalization to unseen applications.
In existing medical entity recognition models,most of them cannot fully extract and utilize the lexical information in the text sequence,and their model structures are complex.This makes these models face problems such as inaccurate entity boundary recognition and poor robustness when dealing with medical texts.Additionally,most word-granularity based named entity recogni-tion(NER)methods are not perfect in solving the problem of information omission.To address these problems,a named entity rec-ognition model based on word fusion and adversarial training is proposed in this paper.The model uses a pre-trained model BERT to obtain word vectors of text sequences.Then the SoftLexicon is used to introduce lexical information and add perturbation samples generated by adversarial training to the word vectors.Finally,the BiLSTM-CRF is used to extract features and obtain sequence anno-tation results.The proposed model is experimented on the datasets CCKS2019 and CCKS2020,where the F1 values reach 85.07%and 90.39%,respectively.The experimental results show that compared with the baseline model,the F1 value of this model has in-creased by 2.31%and 2.88%,indicating that the combination of word fusion method and adversarial training can effectively identify medical entities.
Label-specific features learning is a prominent research hotspot in the field of multi-label learning, which aims to construct a classification model based on the distinctive features of each label rather than the whole features. Existing approaches regarding label-specific features usually assume that the importance of each label to an instance is equal. However, this popular strategy might be suboptimal as the importance of labels actually is different. In this paper, a multi-label-specific features learning algorithm based on label importance and fuzzy rough set is proposed. First, the importance of labels is measured based on the similarity of instances, which not only preserves the ranking of relevant and irrelevant labels, but also follows the principles of smoothness and normalization. Second, the correlation between labels is analyzed, and label-specific features of each label are extracted through a fuzzy rough set model. Experiments on several public available data sets demonstrate the effectiveness of the proposed algorithm.
Structure connectivity and substructure connectivity are two parameters generalized from connectivity. Let G be a graph,H be a certain connected subgraph of G. A set F is called an H-structure cut(resp. H-substructure cut) of G if every element of F is isomorphic to H(resp. isomorphic to connected subgraph of H) and G-Fis disconnected. The H-structure connectivity κ( G; H)(resp. H-substructure connectivity κ s (G; H)) is the cardinality of a minimal H-structure cut(resp. H-substructure cut) of G. In this paper, we determined the Pk structure connectivity κ( FCQ n ; P k ) and substructure connectivity κ s (( FCQ n ; P k )( 3 ≤ k ≤ n) where FCQ n is the n-dimensional folded crossed cube.
The centralized system in ancient China originated from patriarchal cooperative agriculture in the pre-national era. The centralized system of government was established in the Xia, Shang, and Zhou “kingship power” era. In the “imperial power” era from the Qin Dynasty to the Qing Dynasty, the authoritarian centralized system with imperial power as the core was increasingly strengthened. The evolution from “kingship power” to “imperial power” was an inevitable trend of China’s historical development, and the centralized power system had also been strengthened. As a dominant ideology, the thought of “Great Unity” provided a theoretical cornerstone for the centralized system in ancient China; the centralized system in ancient China provided the institutional guarantee for the official status of the idea of “Great Unity.” The system of centralization has been continuous in Chinese history, perfectly adaptable to Chinese traditional society, and interchangeable with the idea of “Great Unity,” providing an ideological basis and institutional guarantee for the formation, continuation, and development of a unified multi-ethnic country in China. The thought of “Great Unity” and centralization are the common values of the Chinese nation, which can provide useful reference for the current road of modernization.