The rapid rise of e-commerce platforms has changed people's shopping habits, driving the popularity online shopping. Users express their opinions on products and services by purchasing posting comments. These comment data contain rich user experiences, which are crucial for enterprises to understand needs improve product quality. Sentiment analysis text is an important research direction in mining, focusing how extract evaluations from provide comprehensive, authentic, accurate feedback....

In the era of big data, major e-commerce platforms are facing challenge an exponential growth in number user comments. Effectively utilizing these comments has become urgent issue. Traditional manual statistical methods no longer able to meet demands for accuracy and real-time analysis. The rise artificial intelligence-based text mining techniques provides a new approach address this problem. By building deep learning analysis models, it is possible uncover preferences product...

With the rapid development of Internet technology, many industries have begun digital transformation. However, while bringing convenience to users, has also become a hotbed for criminals commit fraud. On one hand, large number users on more or less left data, can use this information practice accurate fraud improve success rate fraud; other online financial transactions such as banks and e-commerce provide ways Therefore, all kinds methods emerge in an endless flow, through telephone,...

Visualization of marine environmental field elements is one the core technologies in science research. Particularly context “digital twin ocean” (DTO) construction and application, accurately reproducing dynamic evolution remains a critical challenge. Existing visualization methods are primarily limited to static displays fail achieve deep integration expression sea conditions. To address this, this paper proposes new method for element fields twin-space framework. This first constructs wave...

The rapid rise of e-commerce platforms has changed people's shopping habits, driving the popularity online shopping. Users express their opinions on products and services by purchasing posting comments. These comment data contain rich user experiences, which are crucial for enterprises to understand needs improve product quality. Sentiment analysis text is an important research direction in mining, focusing how extract evaluations from provide comprehensive, authentic, accurate feedback....

Though deep convolutional neural networks (CNNs) have made great breakthroughs in the field of vision-based gesture recognition, however it is challenging to deploy these high-performance resource-constrained mobile platforms and acquire large numbers labeled samples for training CNNs. Furthermore, there are some application scenarios with only a few or even single one new class so that recognition method based on CNNs cannot achieve satisfactory classification performance. In this paper,...

In the age of Internet, online customers express their opinions on products by posting reviews. It is critical to do sentiment analysis customers' review data help subsequent make purchasing decisions and guide companies improve products. Aspect-Category Sentiment Analysis(ACSA) a subtask analysis, which aims detect aspect categories mentioned in text identify corresponding polarities. Most previous methods regard Aspect Category Detection(ACD) AspectCategory Classification(ACSC) as two...

It is important to consider where, when, and how the evolution of sea surface temperature anomalies (SSTA) plays significant roles in regional or global climate changes. In comparison where there a great challenge clearly describing SSTA evolves space time. light from generation, through development, dissipation SSTA, this paper proposes novel approach identifying an time time-series raster dataset. This method, called PoAIES, includes three key steps. Firstly, cluster-based method enhanced...

Network representation learning is receiving increasing attention from scholars. Among them, methods based on graph neural networks have become particularly popular. However, most existing currently only focus with a single type of relations. In the real world, contain wealth diverse information, multiple types relationships between nodes. this paper, we propose network-based multiplex network model (GNMRL). We nodes within each layer by aggregating neighbor information. Additionally, since...

Abstract Model-driven software development has become a hot research topic and discovery trend in the field of engineering. Its core idea is to treat analysis design models as equivalent code. Better integration code can greatly increase chances effective improvement achieve automated through abstract models. In this paper, we first constructed data model-driven architecture system based on meta-modeling hierarchy, using dictionary for storage. The mapping relation loader transforms...

Abstract With the development and popularization of Internet technology, society has put forward higher requirements for form process software. In this study, firstly, based on O-RGPS metamodeling framework, which provides modeling perspectives meta-relationships among them software engineering domain modeling, semantics ontology, role, target models in application are constructed. An extended semantic web is utilized to make network correspond sentence information. Then, conjunction with...

Human beings is entering the Information Age, and E-government has become main direction of administrating innovation for many countries. It information's character geographical space positional relevancy that makes GIS utilize spatial data related technologies play important role in government administration. The paper comprehensively studied on application running mechanism, model frame guarantee GIS.

Multiplex networks contain multiple types of relations between nodes, where each relation type is modeled as one layer. In the real world, a may only depend on certain attribute features nodes. Most existing multiplex network embedding methods focus preserving consistent information or complementary from networks. However, these ignore dependency node attributes and topology relation. To address problem, we propose model called DAME (Disentangled-based Adversarial Network for Embedding). We...

Precipitation extremes driven by the El Niño–Southern Oscillation (ENSO) are one of critical ways in which ENSO impacts global climate, specifically tropical Pacific, where they have potential to cause extreme weather conditions. However, existing approaches struggle effectively identify evolution ENSO-related precipitation anomalies that change rapidly spatial distribution. To address this challenge, we propose object-oriented spatiotemporal clustering approach using remote sensing products...

Many effective and advanced methods have been developed to explore oceanic dynamics using time series of raster-formatted datasets; however, they generally designed at a scale suitable for data observation used independently each other, despite the potential advantages combining different modules into an integrated system suited dynamic evolution. From datasets marine knowledge, we several mining algorithms evolutionary combined them six modules: module dataset pretreatment; process-oriented...