Graph Neural Networks (GNNs) have demonstrated strong capabilities in processing structured data. While traditional GNNs typically treat each feature dimension equally important during graph convolution, we raise an important question: **Is the graph convolution operation equally beneficial for each feature?** If not, the convolution operation on certain feature dimensions can possibly lead to harmful effects, even worse than convolution-free models. Therefore, it is required to distinguish convolution-favored and convolution-disfavored features. Traditional feature selection methods mainly focus on identifying informative features or reducing redundancy, but they are not suitable for structured data as they overlook graph structures. In graph community, some studies have investigated the performance of GNN with respect to node features using feature homophily metrics, which assess feature consistency across graph topology. Unfortunately, these metrics do not effectively align with GNN performance and cannot be reliably used for feature selection in GNNs. To address these limitations, we introduce a novel metric, Topological Feature Informativeness (TFI), to distinguish GNN-favored and GNN-disfavored features, where its effectiveness is validated through both theoretical analysis and empirical observations. Based on TFI, we propose a simple yet effective Graph Feature Selection (GFS) method, which processes GNN-favored and GNN-disfavored features with GNNs and non-GNN models separately. Compared to original GNNs, GFS significantly improves the extraction of useful topological information from each feature with comparable computational costs. Extensive experiments show that after applying GFS to $\textbf{8}$ baseline and state-of-the-art (SOTA) GNN architectures across $\textbf{10}$ datasets, $\textbf{90\%}$ of the GFS-augmented cases show significant performance boosts. Furthermore, our proposed TFI metric outperforms other feature selection methods for GFS. These results verify the effectiveness of both GFS and TFI. Additionally, we demonstrate that GFS's improvements are robust to hyperparameter tuning, highlighting its potential as a universally valid method for enhancing various GNN architectures. To facilitate reproducibility and further research, we have made our code publicly available at https://github.com/KTTRCDL/graph-feature-selection.

Under circumstances of heterophily, where nodes with different labels tend to be connected based on semantic meanings, Graph Neural Networks (GNNs) often exhibit suboptimal performance. Current studies on graph heterophily mainly focus on aggregation calibration or neighbor extension and address the heterophily issue by utilizing node features or structural information to improve GNN representations. In this paper, we propose and demonstrate that the valuable semantic information inherent in heterophily can be utilized effectively in graph learning by investigating the distribution of neighbors for each individual node within the graph. The theoretical analysis is carried out to demonstrate the efficacy of the idea in enhancing graph learning. Based on this analysis, we propose HiGNN, an innovative approach that constructs an additional new graph structure, that integrates heterophilous information by leveraging neighbor distribution to enhance connectivity between nodes that share similar semantic characteristics. We conduct empirical assessments on node classification tasks using both homophilous and heterophilous benchmark datasets and compare HiGNN to popular GNN baselines and SoTA methods, confirming the effectiveness in improving graph representations. In addition, by incorporating heterophilous information, we demonstrate a notable enhancement in classicial GNN baselines across real-world datasets, thus affirming the efficacy of our approach.

Graph homophily refers to the phenomenon that connected nodes tend to share similar characteristics. Understanding this concept and its related metrics is crucial for designing effective Graph Neural Networks (GNNs). The most widely used homophily metrics, such as edge or node homophily, quantify such "similarity" as label consistency across the graph topology. These metrics are believed to be able to reflect the performance of GNNs, especially on node-level tasks. However, many recent studies have empirically demonstrated that the performance of GNNs does not always align with homophily metrics, and how homophily influences GNNs still remains unclear and controversial. Then, a crucial question arises: What is missing in our current understanding of homophily? To figure out the missing part, in this paper, we disentangle the graph homophily into three aspects: label, structural, and feature homophily, which are derived from the three basic elements of graph data. We argue that the synergy of the three homophily can provide a more comprehensive understanding of GNN performance. Our new proposed structural and feature homophily consider the neighborhood consistency and feature dependencies among nodes, addressing the previously overlooked structural and feature aspects in graph homophily. To investigate their synergy, we propose a Contextual Stochastic Block Model with three types of Homophily (CSBM-3H), where the topology and feature generation are controlled by the three metrics. Based on the theoretical analysis of CSBM-3H, we derive a new composite metric, named Tri-Hom, that considers all three aspects and overcomes the limitations of conventional homophily metrics. The theoretical conclusions and the effectiveness of Tri-Hom have been verified through synthetic experiments on CSBM-3H. In addition, we conduct experiments on $31$ real-world benchmark datasets and calculate the correlations between homophily metrics and model performance. Tri-Hom has significantly higher correlation values than $17$ existing metrics that only focus on a single homophily aspect, demonstrating its superiority and the importance of homophily synergy. Our code is available at https://github.com/zylMozart/Disentangle_GraphHom.

Eccentricity detection is a crucial issue for high-speed rotating machinery, which concerns the stability and safety of the machinery. Conventional techniques in industry for eccentricity detection are mainly based on measuring certain physical indicators, which are costly and hard to deploy. In this paper, we propose RED, a non-intrusive, low-cost, and real-time RFID-based eccentricity detection approach. Differing from the existing RFID-based sensing approaches, RED utilizes the temporal and phase distributions of tag readings as effective features for eccentricity detection. RED includes a Markov chain based model called RUM, which only needs a few sample readings from the tag to make a highly accurate and precise judgement. The design of RED further addresses practical issues, such as parameterizing the RUM model, making it robust to dynamic and noisy environments, and considering how the doppler shift may affect our system. We implement RED with COTS RFID reader and tags, and evaluate its performance across various scenarios. The overall accuracy is 93.6 percent and the detection latency is 0.68 seconds in average.

With parallel decoding for backscatter communication, tags are allowed to transmit concurrently and more efficiently. Existing parallel decoding mechanisms, however, assume that signals of the tags are highly stable and, hence, may not perform optimally in the naturally dynamic backscatter systems. This paper introduces FlipTracer, a practical system that achieves highly reliable parallel decoding even in hostile channel conditions. FlipTracer is designed with a key insight; although the collided signal is time-varying and irregular, transitions between signals' combined states follow highly stable probabilities, which offers important clues for identifying the collided signals and provides us with an opportunity to decode the collided signals without relying on stable signals. Motivated by this observation, we propose a graphical model, called one-flip-graph (OFG), to capture the transition pattern of collided signals, and design a reliable approach to construct the OFG in a manner robust to the diversity in backscatter systems. Then, FlipTracer can resolve the collided signals by tracking the OFG. We have implemented FlipTracer and evaluated its performance with extensive experiments across a wide variety of scenarios. Our experimental results have shown that FlipTracer achieves a maximum aggregated throughput that approaches 2 Mb/s, which is 6× higher than the state of the art.

Eccentricity detection is a crucial issue for highspeed rotating machinery, which concerns the stability and safety of the machinery. Conventional techniques in industry for eccentricity detection are mainly based on measuring certain physical indicators, which are costly and hard to deploy. In this paper, we propose RED, a non-intrusive, low-cost, and realtime RFID-based eccentricity detection approach. Differing from the existing RFID-based sensing approaches, RED utilizes the temporal and phase distributions of tag readings as effective features for eccentricity detection. RED includes a Markov chain based model called RUM, which only needs a few sample readings from the tag to make a highly accurate and precise judgement. We implement RED with commercial-of-the-shelf RFID reader and tags, and evaluate its performance across various scenarios. The overall accuracy is 93.59% and the detection latency is 0.68 seconds in average.

With parallel decoding for backscatter communication, tags are allowed to transmit concurrently and more efficiently. Existing parallel decoding mechanisms, however, assume that signals of the tags are highly stable, and hence may not perform optimally in the naturally dynamic backscatter systems. This paper introduces FlipTracer, a practical system that achieves highly reliable parallel decoding even in hostile channel conditions. FlipTracer is designed with a key insight: although the collided signal is time-varying and irregular, transitions between signals' combined states follow highly stable probabilities, which offers important clues for identifying the collided signals, and provides us with an opportunity to decode the collided signals without relying on stable signals. Motivated by this observation, we propose a graphical model, called one-flip-graph (OFG), to capture the transition pattern of collided signals, and design a reliable approach to construct the OFG in a manner robust to the diversity in backscatter systems. Then FlipTracer can resolve the collided signals by tracking the OFG. We have implemented FlipTracer and evaluated its performance with extensive experiments across a wide variety of scenarios. Our experimental results have shown that FlipTracer achieves a maximum aggregated throughput that approaches 2 Mbps, which is 6x higher than the state-of-the-art.