Most of the traditional outlier mining algorithms are not suitable for high dimensional datasets. However, the dimensionality of the dataset in the field of the Internet of Things is high. Therefore, Hypergraph-based ...
详细信息
Sequential pattern mining has a wide range of applications in data streams. The real data involves multiple data streams and each data stream is itemset-sequence. However, most algorithms mine a single item in a singl...
详细信息
Based on the centralized channel allocation strategy in the Cognitive Radio Networks, considering the opportunistic channel access for Secondary Users, a discretetime queueing model with a general transmission time an...
详细信息
In this paper, we proposed a new sequential pattern mining algorithm called WSPD for mining weighted sequential patterns in data streams. The algorithm produces no false negatives and places a bound on the error of th...
详细信息
Consider traditional clustering algorithms seldom had a research on users’ visit behavior and content, and they cannot cluster users with similar visit behavior into a community easily. Behavior of user cannot cluste...
详细信息
With the explosive growth of wireless application, how to improve the spectrum efficiency as well as reduce the communication consumption is a hot topic of research. In this paper, we propose a novel energy saving str...
详细信息
Software behavior mining is a very meaningful work. Finding that desirable patterns can assist the program maintainers to comprehend the software adequately. Although the existing high utility pattern mining algorithm...
详细信息
In this paper, we present a new method for constructing quaternary sequence with even period 2f based on cyclotomic classes of order two. Under the premise of v = 2f +1, we construct two binary sequences with period 2...
详细信息
Maximal frequent sequence mining is an important research issue which has realized the highly compressed storage of frequent sequences. At present, most algorithms are based on bottom-up method and large numbers of ca...
详细信息
Vision transformers have significantly advanced the field of computer vision in recent years. The cornerstone of these transformers is the multi-head attention mechanism, which models interactions between visual eleme...
Vision transformers have significantly advanced the field of computer vision in recent years. The cornerstone of these transformers is the multi-head attention mechanism, which models interactions between visual elements within a feature map. However, the vanilla multi-head attention paradigm independently learns parameters for each head, which ignores crucial interactions across different attention heads and may result in redundancy and under-utilization of the model’s capacity. To enhance model expressiveness, we propose a novel nested attention mechanism, Ne-Att, that explicitly models cross-head interactions via a hierarchical variational distribution. We conducted extensive experiments on image classification, and the results demonstrate the superiority of Ne-Att.
暂无评论