Abstract This paper proposes an attention-based LSTM (AT-LSTM) model for financial time series prediction. We divide the prediction process into two stages. For first stage, we apply attention to assign different weights input features of at each step. In second feature is utilized effectively select relevant sequences as neural network in next frame. Our proposed framework not only solves long-term dependence problem effectively, but also improves interpretability methods based on network....

Submodular maximization has found extensive applications in various domains within the field of artificial intelligence, including but not limited to machine learning, computer vision, and natural language processing. With increasing size datasets these domains, there is a pressing need develop efficient parallelizable algorithms for submodular maximization. One measure parallelizability algorithm its adaptive complexity, which indicates number sequential rounds where polynomial queries...

Given the edge list of a social network, node embedding method learns structural features for every and embeds into vector space. The current related work on exploits only portion existing networks, e.g., static networks. However, networks are inherently hierarchical dynamic systems in which topology changes constantly strength influence information among neighbors varies with different numbers hops. We propose highly efficient method, DNPS, that is faster more accurate than state-of-the-art...

Submodular maximization has wide applications in machine learning and data mining, where massive datasets have brought the great need for designing efficient parallelizable algorithms. One measure of parallelizability a submodular algorithm is its adaptivity complexity, which indicates number sequential rounds polynomial queries to objective function can be executed parallel. In this paper, we study problem non-monotone subject knapsack constraint, propose first combinatorial achieving an...

We present an algorithm (LsNet2Vec) that, given a large-scale network (millions of nodes), embeds the structural features node into lower and fixed dimensions vector in set real numbers. experiment evaluate our proposed approach with twelve datasets collected from SNAP. Results show that model performs comparably state-of-the-art methods, such as Katz method Random Walk Restart method, various settings.

Learning distributed representations of symbolic data were introduced by Hinton[1], and first developed in modeling networks for learning the node vectors Perozzi et al (2014). In this work, we proposed Dnps, a novel nodes embedding approach acquiring large-scale dynamic social networks. Dnps is suitable many types networks: dynamic/static, directed/undirected, weighted/unweighted. Recently, several works proposed. However, they designed static networks, such as language To address problem,...

In this paper, we present Cdear, a novel system to detect fake reviews by using sentiment analysis on attributes of products. We formulate review spam detection as an opinion coincided problem. Specifically, try capture the diverse products among different consumers. To our knowledge, is very first attempt use in spam. evaluate effectiveness system, conduct experiments real life datasets and employ 20 experts assess reliability carrying out simulation shopping online. Meanwhile, developed...

Enterprise competition analysis has long been treated as a crucial task of management science, which can reveal pertinent information about market saturation and business opportunities to support the decision-making process entrepreneurs investors. Recently, with development graph representation techniques, enterprises could be now formulated in novel graph-oriented perspective model their cooperation directional graph. However, these prior arts mainly treat individual nodes, while...

With the continuous development and change exhibited by large language model (LLM) technology, represented generative pretrained transformers (GPTs), many classic scenarios in various fields have re-emerged with new opportunities. This paper takes ChatGPT as modeling object, incorporates LLM technology into typical book resource understanding recommendation scenario for first time, puts it practice. By building a ChatGPT-like system (BookGPT) framework based on ChatGPT, this attempts to...

Submodular maximization has found extensive applications in various domains within the field of artificial intelligence, including but not limited to machine learning, computer vision, and natural language processing. With increasing size datasets these domains, there is a pressing need develop efficient parallelizable algorithms for submodular maximization. One measure parallelizability algorithm its adaptive complexity, which indicates number sequential rounds where polynomial queries...

This paper studies the problem of learning node embeddings (a.k.a. distributed representations) for dynamic networks. The embedding methods allocate each in network with a d-dimensions vector, which can generalize across various tasks, such as item recommendation, labeling, and link prediction. In practice, many real-world networks are evolving nodes/links added or deleted. However, most proposed focusing on static Although some previous researches have shown promising results to handle...