Accurate traffic prediction is a critical yet challenging task in Intelligent Transportation Systems, benefiting variety of smart services, e.g., route planning and management. Although extensive efforts have been devoted to this problem, it still not well solved due the flexible dependency within data along both spatial temporal dimensions. In paper, we explore flexibility from three aspects, namely time-varying local dependency, dynamic global dependency. Then propose novel Dual Graph...

Depression affects more than 300 million people around the world and is leading cause of disability in USA for individuals ages from 15 to 44. The damage it compares most common diseases like cancer, diabetes, or heart disease according WHO report. However, with depression symptoms sometimes do not receive proper treatment due access barriers. In this paper, we propose a method that automatically detects using only landmarks facial expressions, which are easy collect less privacy exposure....

Chart understanding enables automated data analysis for humans, which requires models to achieve highly accurate visual comprehension. While existing Visual Language Models (VLMs) have shown progress in chart understanding, the lack of high-quality training and comprehensive evaluation benchmarks hinders VLM In this paper, we introduce EvoChart, a novel self-training method generating synthetic enhance VLMs' capabilities real-world We also propose EvoChart-QA, noval benchmark measuring...

To achieve low-cost and robustness, an indoor location system using simple visual tags is designed by comprehensively considering accuracy computation complexity. Only the color shape features are used for tag detection, which both algorithm complexity data storage requirement reduced. manage nonunique problem caused features, a fast query matching method further presented view field of camera azimuth. Then, based on relationship analysis between spatial distribution error, pose position...

As a neurodegenerative disease, Parkinson's disease (PD) is hard to identify at the early stage, while using speech data build machine learning diagnosis model has proved effective in its diagnosis. However, show high degrees of redundancy, repetition, and unnecessary noise, which influence accuracy results. Although feature reduction (FR) could alleviate this issue, traditional FR one-sided (traditional extraction construct high-quality features without preference, selection achieve...

Integrating the health preserving concept of Traditional Chinese medicine (TCM) into recommendation system can provide users with personalized and healthy information push. By constructing an ontology-based TCM model in ordering system, we designed context rules that conform to concept, by iteratively improving user information, based on demographic recommendations, improved status sparseness tendency singular recommendations recommend similarity User Profiling context.

Motion planning by considering it as an optimal problem is effective and widely applicable method. Its comprehensive performance greatly depends on the vehicle dynamics model, which highly coupled nonlinear, especially under dynamical scenarios causes much more consumption of computation resources for numerical optimization. To increase real time motion planner designed nonlinear model predictive control (NMPC), a unified simplified (SDM) presented to make balance between accuracy complexity...

In this paper, we consider the following viscoelastic equations with initial condition and zero Dirichlet boundary condition. Using concavity method, obtained sufficient conditions on data arbitrarily high energy such that solution blows up in finite time.

The applications of parallel robots in various fields are increasing. In this paper, a research on the design optimization 6-PSS manipulator for motion simulating is presented. problem maximizing workspace formulated according to geometric and kinematic analysis. particular, totally flexible with 6 degrees-of-freedom (DOF). To maximize robot, we investigated differential evolution algorithm (DE), genetic (GA) also adopted as comparison. Computational results show DE outperforms GA at about...

In this paper, we consider the long time behavior of solutions initial value problem for viscoelastic wave equation under boundary damping \begin{eqnarray*} u_{tt}-\Delta u+\int_0^t g(t-\tau)\text{div}(a(x)\nabla u(\tau))d\tau+u_t=0 &\text{in}\,\Omega\times(0,\infty). \end{eqnarray*} For low energy case, which is non-positive energy, based on concavity argument prove blow up result. As high give out sufficient conditions datum such that solution blows in finite time.

Chronic diseases are severe and life-threatening, their accurate early diagnosis is difficult. Machine-learning-based processes of data collected from the human body using wearable sensors a valid method currently usable for diagnosis. However, it difficult sensor systems to obtain high-quality large amounts meet demands diagnostic accuracy. Furthermore, existing feature-learning methods do not deal with this problem well. To address above issues, sample-pair envelope diamond autoencoder...

Chart understanding enables automated data analysis for humans, which requires models to achieve highly accurate visual comprehension. While existing Visual Language Models (VLMs) have shown progress in chart understanding, the lack of high-quality training and comprehensive evaluation benchmarks hinders VLM In this paper, we introduce EvoChart, a novel self-training method generating synthetic enhance VLMs' capabilities real-world We also propose EvoChart-QA, noval benchmark measuring...

The current cold-formed steel (CFS) wall is limited to the low-rise thin-walled structure, which inappropriate for multi-story constructions due its low shear resistance and poor corrosion of outer sheathing board. Therefore, a new composite CFS frame with concrete plasterboard was proposed, cast-in situ layer used surface in response needs resistance, utilized inner reduce weight. This paper presented quasi-static test on seismic behavior one three composite-walls’ sheathings obtain...