Generating the precise semantic representation of a word/concept is a fundamental task in natural language processing. Recent studies which incorporate semantic knowledge into word embedding have shown their potential in improving the semantic representation of a concept. However, existing approaches only achieved limited performance improvement as they usually (1) model a word's semantics from some explicit aspects while ignoring the intrinsic aspects of the word, (2) treat semantic knowledge as a supplement of word embeddings, and (3) consider partial relations between concepts while ignoring rich coupling relations between them, such as explicit concept co-occurrences in descriptive texts in a corpus as well as concept hyperlink relations in a knowledge network, and implicit couplings between the explicit relations. In human consciousness, concepts are associated with various coupling relations, which inspires us to capture as many concept couplings as possible for building a better concept representation. We thus propose a neural coupled concept representation (CoupledCR) framework and its instantiation: a coupled concept embedding (CCE) model. CCE first learns two types of explicit couplings from concept cooccurrences and hyperlink relations respectively, and then learns a type of high-level implicit couplings between these two types of explicit couplings. Extensive experimental results on real-world datasets show that CCE significantly outperforms state-of-the-art semantic representation methods.

Community detection is a challenging task in attributed networks, due to the data inconsistency between network topological structure and node attributes. The problem of how to effectively and robustly fuse multi-source heterogeneous data plays an important role in community detection algorithms. Although some algorithms taking both topological structure and node attributes into account have been proposed in recent years, the fusion strategy is simple and usually adopts the linear combination method. As a consequence of this, the detected community structure is vulnerable to small variations of the input data. In order to overcome this challenge, we develop a novel two-layer representation to capture the latent knowledge from both topological structure and node attributes in attributed networks. Then, we propose a weighted co-association matrix-based fusion algorithm (WCMFA) to detect the inherent community structure in attributed networks by using multi-layer fusion strategies. It extends the community detection method from a single-view to a multi-view style, which is consistent with the thinking model of human beings. Experiments show that our method is superior to the state-of-the-art community detection algorithms for attributed networks.

As a foundational analytical tool, the discernibility matrix plays a pivotal role in the systematic reduction of knowledge in rough set-based systems. Recent advancements in rough set theory have witnessed the proliferation of discernibility matrix-based knowledge reduction algorithms, with notable applications in classical, neighborhood, covering, and fuzzy rough set models. However, the quadratic growth of the discernibility matrix’s complexity (relative to domain size) imposes fundamental scalability limits, rendering it inefficient for real-world applications with massive datasets. To address this issue, we introduced a discernibility hashing strategy to limit the growth scale of the discernibility attributes and proposed a feature selection algorithm via discernibility hash based on rough set theory. First, on the premise of keeping the information of the original discernibility matrix unchanged, the method maps the discernibility attribute set of all objects to the storage unit through a hash function and records the number of collisions to construct a discernibility hash. By using this mapping, the two-dimensional matrix space can be reduced to a one-dimensional hash space, which greatly removes invalid and redundant elements. Secondly, based on the discernibility hash, an efficient knowledge reduction algorithm is proposed. The algorithm avoids invalid and redundant element attribute sets to participate in the knowledge reduction process and improves the efficiency of the algorithm. Finally, the experimental results show that the method is superior to the discernibility matrix method in terms of storage space and running time.