Bipartite Matching Algorithm Research Articles

Multimodal Data Mining Based on Self Attention Feature Alignment

This study explores the application of multimodal data mining in text and image vector processing, aiming to improve the depth and breadth of data analysis by integrating information from different data types. We use the FairFace dataset combined with the CLIP model encoding layer to obtain text and image vectors, and use the K-Means clustering algorithm to achieve vector dimensionality reduction. Subsequently, we introduced the bipartite graph matching algorithm to achieve maximum matching between text vectors and image vectors, and calculated the contrastive learning loss and similarity loss. The entire process covers steps such as data preparation, feature extraction, vector dimensionality reduction, matching algorithms, and loss assessment, constructing a complete text and image matching task process. Our research contributions include using K-Means clustering algorithm to achieve vector dimensionality reduction, as well as introducing bipartite graph matching algorithm and calculating two types of losses in text and image vector matching, further improving matching quality.

Trajectory privacy protection method with smart contract-based query exchange in the Social Internet of Vehicles

Query exchange in the Social Internet of Vehicles (SIoV) can protect users’ trajectory information. However, this method lacks an appropriate incentive mechanism, which leads to cooperative users refusing to participate in query exchange. In order to provide cooperative users with incentives to participate in query exchange, this paper proposes a smart contract-based query exchange (SC-QE) trajectory privacy protection method. By creating a many-to-many smart contract, the method encourages the cooperative users to bid to the requesting users. Subsequently, in order to select a Best Similarity Deviation User (BSDU) for the requesting user to perform query exchange, the users in the smart contract are modeled as a weighted bipartite graph, and the matching between the requesting users and BSDUs is realized by means of a weighted bipartite graph best matching algorithm. Following successful verification of the query exchange transaction in the smart contract, the base station distributes rewards to the BSDU and uploads the query exchange transaction to the consortium blockchain. Experimental results show that compared with the deviation-based query exchange (DQE) method, the proposed method reduces the user processing time by 12% while increasing the continuous anonymous success rate by 29%. Therefore, the proposed method can reduce the service query time and improve the level of trajectory privacy protection.

Utility-Aware Dynamic Ridesharing in Spatial Crowdsourcing

With smartphones and geo-locating devices widely used around the world, ridesharing, as a main application field of spatial crowdsourcing, has been fast expanding its widespread adoption and potentially brings great benefits to human society and the environment. However, ridesharing has so far not been as popular as expected. A recent survey shows that notable obstacles come from concerns about social comfort and price fairness when riding with strangers. To defeat these obstacles, in this paper, we study a novel problem of utility-aware ride matching (URM) for dynamic ridesharing, where drivers and riders are matched in batches by considering their social comfort and price revenue. While the URM problem is of practical usefulness, we prove that this problem is Nondeterministic Polynomial Hard (NP-hard). To tackle the problem optimally and find exact answers, we present a novel bipartite matching algorithm by integrating an effective Driver-Rider Graph (DR-Graph) index. To balance the effectiveness and efficiency, we propose two efficient algorithms to solve the URM problem with only a small loss of utility. Leveraging a bounded Driver-Rider-group Graph (DRg-Graph) and several useful pruning bounds for matching utility and travel cost, we develop an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\epsilon$</tex-math></inline-formula> -refining algorithm to find an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\epsilon$</tex-math></inline-formula> -approximation matching utility of the optimal answer. Extensive experiments on real datasets showed that our proposed algorithms achieved effective matching results of social-pricing based utilities and the efficient performance of quick matching under various parameter settings.

Online Bipartite Matching with Reusable Resources

We study the classic online bipartite matching problem with a twist: off-line vertices, called resources, are reusable. In particular, when a resource is matched to an online vertex, it is unavailable for a deterministic time duration d, after which it becomes available again for a rematch. Thus, a resource can be matched to many different online vertices over a period of time. Whereas recent work on the problem has resolved the asymptotic case in which we have large starting inventory (i.e., many copies) of every resource, we consider the (more general) case of unit inventory and give the first algorithms that are provably better than the naïve greedy approach, which has a competitive ratio of (exactly) 0.5. Our first algorithm, which achieves a competitive ratio of 0.589, generalizes the classic RANKING algorithm for online bipartite matching of nonreusable resources (Karp et al. 1990) by reranking resources independently over time. Whereas reranking resources frequently has the same worst case performance as greedy, we show that reranking intermittently on a periodic schedule succeeds in addressing reusability of resources and performs significantly better than greedy in the worst case. Our second algorithm, which achieves a competitive ratio of 0.505, is a primal-dual randomized algorithm that works by suggesting up to two resources as candidate matches for every online vertex and then breaking the tie to make the final matching selection in a randomized correlated fashion over time. As a key component of our algorithm, we suitably adapt and extend the powerful technique of online correlated selection (Fahrbach et al. 2020) to reusable resources in order to induce negative correlation in our tie-breaking step and beat the competitive ratio of 0.5. Both of our results also extend to the case in which off-line vertices have weights. Funding: R. Niazadeh’s research is supported by the Asness Junior Faculty Fellowship at The University of Chicago Booth School of Business.

Efficacy of oral SERDs in the treatment of ER+, HER2− metastatic breast cancer: a stratified analysis of the ESR1 wild type and mutant subgroups

Oral SERDs are a novel drug class that have been developed to counteract resistance due to ESR1 mutations. Several SERDs have emerged from phase 2 and 3 trials, with the FDA limiting approval for Elacestrant to patients with ESR1mt tumours despite PFS benefit in the overall population. However, questions remain on whether patients with ESR1wt tumours stand to benefit from oral SERDs. Manuscripts and conference presentations of Randomised Controlled Trials were extracted after a systematic search of Embase, PubMed and Cochrane from inception until January 21,2023. RCTs investigating the efficacy of oral SERDs versus endocrine therapy for ER positive, HER2 negative advanced breast cancer, and which reported the Kaplan Meier (KM) curves of PFS in the overall and ESR1 mutant (ESR1mt) population were selected. A graphical reconstructive algorithm was applied to estimate time-to-event outcomes from reported KM curves in all overall and ESR1mt cohorts. A bipartite matching algorithm, KMSubtraction, was used to derive survival data for unreported (ESR1wt) subgroups. An individual patient data (IPD) meta-analysis was then pursued, pooling data by ESR1 mutation status in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Cochrane Guidelines for IPD. The randomized clinical trials ACELERA, AMEERA-3, EMERALD and SERENA-2 were included, totalling 1290 patients. In the pooled analysis of the overall cohort, PFS benefit was observed with oral SERDs when compared with treatment of physicians choice (TPC) (HR 0.783, 95%CI 0.681-0.900, p<0.001). In the ESR1mt subgroup, oral SERDs demonstrated improved PFS (HR 0.557, 95%CI 0.440-0.705, p<0.001) compared to TPC. In the ESR1wt subgroup, oral SERDs demonstrated no significant PFS benefit (HR 0.944, 95%CI 0.783-1.138, p=0.543) when compared to TPC. The results of this IPD meta-analysis suggests that PFS benefit in the overall population is mainly driven by the ESR1mt subgroup.

End-to-End Zero-Shot HOI Detection via Vision and Language Knowledge Distillation

Most existing Human-Object Interaction (HOI) Detection methods rely heavily on full annotations with predefined HOI categories, which is limited in diversity and costly to scale further. We aim at advancing zero-shot HOI detection to detect both seen and unseen HOIs simultaneously. The fundamental challenges are to discover potential human-object pairs and identify novel HOI categories. To overcome the above challenges, we propose a novel End-to-end zero-shot HOI Detection (EoID) framework via vision-language knowledge distillation. We first design an Interactive Score module combined with a Two-stage Bipartite Matching algorithm to achieve interaction distinguishment for human-object pairs in an action-agnostic manner. Then we transfer the distribution of action probability from the pretrained vision-language teacher as well as the seen ground truth to the HOI model to attain zero-shot HOI classification. Extensive experiments on HICO-Det dataset demonstrate that our model discovers potential interactive pairs and enables the recognition of unseen HOIs. Finally, our method outperforms the previous SOTA under various zero-shot settings. Moreover, our method is generalizable to large-scale object detection data to further scale up the action sets. The source code is available at: https://github.com/mrwu-mac/EoID.

Efficacy of oral selective estrogen receptor degraders (SERD)s in the treatment of estrogen receptor positive (ER+), HER2-negative metastatic breast cancer (MBC): A stratified analysis of the ESR1 wild type (wt) and mutant (mt) subgroups.

1096 Background: Oral SERDs are a novel drug class developed to counteract endocrine resistance due to ESR1 mutations. Several SERDs have emerged from phase 2 and 3 trials, and the FDA has most recently limited approval for Elacestrant to patients with ESR1mt tumours despite PFS benefit in the overall population. Questions remain on whether patients with ESR1wt tumours stand to benefit from SERDs. We conducted an IPD meta-analysis of the trials EMERALD, SERENA2, ACELERA and AMEERA3 to compare survival outcomes of SERDs in the 2nd or 3rd line setting against endocrine therapy (ET), using KMSubtraction to unveil HRs of unreported ESR1wt subgroups from AMEERA3 and EMERALD. Meta-analysis results were stratified by ESR1 status. Methods: RCTs investigating the efficacy of novel SERDs versus ET for ER+, HER2- MBC, and which reported the Kaplan Meier (KM) plots of progression free survival (PFS) were selected after a systematic search of Embase and PubMed from inception until January 21,2023. A graphical reconstructive algorithm was applied to estimate time-to-event outcomes from reported KM plots in all overall and reported subgroup cohorts, deriving IPD. A bipartite matching algorithm, KMSubtraction was used to subtract patients in the reported subgroup from the overall cohort, to derive survival data for unreported ESR1wt subgroups. An IPD meta-analysis was then performed, pooling data by ESR1 mutation. Results: A total of 1216 patients were included in this analysis. In the pooled analysis of the overall cohort, PFS benefit was observed with SERDs compared to treatment of physician's choice (TPC) (HR 0.787, 95%CI 0.683-0.908, p&lt;0.001). In the ESR1mt subgroup SERDs demonstrated improved PFS (HR 0.557, 95%CI 0.440-0.705, p&lt;0.001) compared with TPC. In the ESR1wt subgroup, SERDs demonstrated no significant PFS benefit (HR 0.948, 95%CI 0.790-1.137, p=0.565) when compared to TPC. KMSubtraction was implemented on AMEERA3 to reveal survival data in ESR1wt subgroup. Amcenestrant revealed no significant PFS difference in the ESR1wt population (HR 1.197, 95% CI 0.857-1.670, p=0.291) when compared to ET. Additionally, a separate analysis of EMERALD was performed using KMSubtraction to derive survival data of the ESR1wt subgroup comparing Elacestrant vs Fulvestrant. No significant PFS difference was observed between the 2 agents in the ESR1wt population (HR 0.832, 95% CI 0.587-1.178, p=0.299). Conclusions: Our results suggest that PFS benefit in the overall population is mainly driven by the ESR1mt subgroup. These findings are in line with the recent FDA approval of Elacestrant for patients with ESR1mt tumours. [Table: see text]

Matching single cells across modalities with contrastive learning and optimal transport.

Understanding the interactions between the biomolecules that govern cellular behaviors remains an emergent question in biology. Recent advances in single-cell technologies have enabled the simultaneous quantification of multiple biomolecules in the same cell, opening new avenues for understanding cellular complexity and heterogeneity. Still, the resulting multimodal single-cell datasets present unique challenges arising from the high dimensionality and multiple sources of acquisition noise. Computational methods able to match cells across different modalities offer an appealing alternative towards this goal. In this work, we propose MatchCLOT, a novel method for modality matching inspired by recent promising developments in contrastive learning and optimal transport. MatchCLOT uses contrastive learning to learn a common representation between two modalities and applies entropic optimal transport as an approximate maximum weight bipartite matching algorithm. Our model obtains state-of-the-art performance on two curated benchmarking datasets and an independent test dataset, improving the top scoring method by 26.1% while preserving the underlying biological structure of the multimodal data. Importantly, MatchCLOT offers high gains in computational time and memory that, in contrast to existing methods, allows it to scale well with the number of cells. As single-cell datasets become increasingly large, MatchCLOT offers an accurate and efficient solution to the problem of modality matching.

An analysis of one-to-one matching algorithms for entity resolution

AbstractEntity resolution (ER) is the task of finding records that refer to the same real-world entities. A common scenario, which we refer to as Clean-Clean ER, is to resolve records across two clean sources (i.e., they are duplicate-free and contain one record per entity). Matching algorithms for Clean-Clean ER yield bipartite graphs, which are further processed by clustering algorithms to produce the end result. In this paper, we perform an extensive empirical evaluation of eight bipartite graph matching algorithms that take as input a bipartite similarity graph and provide as output a set of matched records. We consider a wide range of matching algorithms, including algorithms that have not previously been applied to ER, or have been evaluated only in other ER settings. We assess the relative performance of these algorithms with respect to accuracy and time efficiency over ten established real-world data sets, from which we generated over 700 different similarity graphs. Our results provide insights into the relative performance of these algorithms and guidelines for choosing the best one, depending on the data at hand.

Secrecy rate maximization for 6G cognitive satellite-UAV networks

To cover remote areas where terrestrial cellular networks may not be available, non-terrestrial infrastructures such as satellites and unmanned aerial vehicles (UAVs) can be utilized in the upcoming sixth-generation (6G) era. Considering the spectrum scarcity problem, satellites and UAVs need to share the spectrum to save costs, leading to a cognitive satellite-UAV network. Due to the openness of both satellite links and UAV links, communication security has become a major concern in cognitive satellite-UAV networks. In this paper, we safeguard a cognitive satellite-UAV network from a physical layer security (PLS) perspective. Using only the slowly-varying large-scale channel state information (CSI), we jointly allocate the transmission power and subchannels to maximize the secrecy sum rate of UAV users. The optimization problem is a mixed integer nonlinear programming (MINLP) problem with coupling constraints. We propose a heuristic algorithm which relaxes the coupling constraints by the penalty method and obtains a sub-optimal low-complexity solution by utilizing random matrix theory, the max-min optimization tool, and the bipartite graph matching algorithm. The simulation results corroborate the superiority of our proposed scheme.

Edge-Weighted Online Bipartite Matching

Online bipartite matching is one of the most fundamental problems in the online algorithms literature. Karp, Vazirani, and Vazirani (STOC 1990) gave an elegant algorithm for unweighted bipartite matching that achieves an optimal competitive ratio of 1-1/e . Aggarwal et al. (SODA 2011) later generalized their algorithm and analysis to the vertex-weighted case. Little is known, however, about the most general edge-weighted problem aside from the trivial 1/2-competitive greedy algorithm. In this article, we present the first online algorithm that breaks the long-standing 1/2 barrier and achieves a competitive ratio of at least 0.5086. In light of the hardness result of Kapralov, Post, and Vondrák (SODA 2013), which restricts beating a 1/2 competitive ratio for the more general monotone submodular welfare maximization problem, our result can be seen as strong evidence that edge-weighted bipartite matching is strictly easier than submodular welfare maximization in an online setting. The main ingredient in our online matching algorithm is a novel subroutine called online correlated selection (OCS), which takes a sequence of pairs of vertices as input and selects one vertex from each pair. Instead of using a fresh random bit to choose a vertex from each pair, the OCS negatively correlates decisions across different pairs and provides a quantitative measure on the level of correlation. We believe our OCS technique is of independent interest and will find further applications in other online optimization problems.

Controllability analysis of the small-world network of neural populations

Controllability analysis of brain networks is the theoretical foundation for neuromodulation feasibility. This paper presents a new framework for studying controllability of certain brain networks on the basis of neural mass models, the minimum driver node, the linearization technique and a controllability index. Firstly, a WS small-world network of Jansen-Rit's neural populations is established to mathematically model complicated neural dynamics. Secondly, an analytical method of analyzing controllability is built based on the bipartite graph maximum matching algorithm, the linearization technique and the matrix condition number. The bipartite graph maximum matching algorithm is applied to determine the minimum driver node sets for the established network while the matrix condition number is applied to define the controllability index which qualitatively evaluates the degree of the controllability of the established network. Finally, the effectiveness of the proposed analytical method is demonstrated by the influence of important parameters on the controllability and the comparison with an existing method. The proposed framework provides theoretical foundation for the study of neuromodulation feasibility, and the results are expected to lead us to better modulate neurodynamics by optimizing network dynamics or designing optimal stimulation protocols.

Efficient radio resource allocation scheme for 5G networks with device-to-device communication

&lt;p&gt;A vital technology in the next-generation cellular network is device-to-device (D2D) communication. Cellular user enabled with D2D communication provides high spectral efficiency and further increases the coverage area of the cell, especially for the end-cell users and blind spot areas. However, the implementation of D2D communication increases interference among the cellular and D2D users. In this paper, we proposed a radio resource allocation (RRA) algorithm to manage the interference using fractional frequency reuse (FFR) scheme and Hungarian algorithm. The proposed algorithm is divided into three parts. First, the FFR scheme allocates different frequency bands among the cell (inner and outer region) for both the cellular and the D2D users to reduce the interference. Second, the Hungarian weighted bipartite matching algorithm is used to allocate the resources to D2D users with the minimum total system interference, while maintaining the total system sum rate. The cellular users share the resources with more than one D2D pair. Lastly, the local search technique of swapping is used for further allocation to minimize the interference. We implemented two types of assignments, fair multiple assignment, and restricted multiple assignment. We compared our results with existing algorithms which verified that our proposed algorithm provides outstanding results in aspects like interference reduction and system sum rate. For restricted multiple assignment, 60-70% of the D2D users are allocated in average cases.&lt;/p&gt;

PKSPS: a novel method for predicting kinase of specific phosphorylation sites based on maximum weighted bipartite matching algorithm and phosphorylation sequence enrichment analysis.

With the development of biotechnology, a large number of phosphorylation sites have been experimentally confirmed and collected, but only a few of them have kinase annotations. Since experimental methods to detect kinases at specific phosphorylation sites are expensive and accidental, some computational methods have been proposed to predict the kinase of these sites, but most methods only consider single sequence information or single functional network information. In this study, a new method Predicting Kinase of Specific Phosphorylation Sites (PKSPS) is developed to predict kinases of specific phosphorylation sites in human proteins by combining PKSPS-Net with PKSPS-Seq, which considers protein-protein interaction (PPI) network information and sequence information. For PKSPS-Net, kinase-kinase and substrate-substrate similarity are quantified based on the topological similarity of proteins in the PPI network, and maximum weighted bipartite matching algorithm is proposed to predict kinase-substrate relationship. In PKSPS-Seq, phosphorylation sequence enrichment analysis is used to analyze the similarity of local sequences around phosphorylation sites and predict the kinase of specific phosphorylation sites (KSP). PKSPS has been proved to be more effective than the PKSPS-Net or PKSPS-Seq on different sets of kinases. Further comparison results show that the PKSPS method performs better than existing methods. Finally, the case study demonstrates the effectiveness of the PKSPS in predicting kinases of specific phosphorylation sites. The open source code and data of the PKSPS can be obtained from https://github.com/guoxinyunncu/PKSPS.

WiFED Mobile: WiFi Friendly Energy Delivery With Mobile Distributed Beamforming

Wireless RF energy transfer for indoor sensors is an emerging paradigm ensuring continuous operation without battery limitations. However, high power radiation within ISM band interferes with packet reception for existing WiFi devices. The paper proposes the first effort in merging RF energy transfer within a standards compliant 802.11 protocol, realizing practical and WiFi-friendly Energy Delivery with Mobile Transmitters (WiFED Mobile). WiFED Mobile architecture is composed of a centralized controller coordinating the actions of multiple energy transmitters (ETs), and deployed sensors that periodically requires charging. The paper first describes 802.11 supported protocol features that can be exploited by sensors to request energy and for ETs to participate in energy transfer. Second, it devises a controller-driven bipartite matching algorithm, assigning appropriate number of ETs to sensors for efficient energy delivery. Thirdly, it detects outlier sensors (OS), which have limited power reception from static ETs and utilizes mobile ETs (METs) to satisfy their charging cycles. The proposed in-band and protocol supported coexistence in WiFED Mobile is validated via simulations and partly in a software defined radio testbed, showing that METs reduce latency by 42% and improve throughput by 83% in scenarios where using only static ETs fails to satisfy charging cycles of OS.

Sliding-Window QPS (SW-QPS)

In this work, we first propose a parallel batch switching algorithm called Small-Batch Queue-Proportional Sampling (SB-QPS). Compared to other batch switching algorithms, SB-QPS significantly reduces the batch size without sacrificing the throughput performance and hence has much lower delay when traffic load is light to moderate. It also achieves the lowest possible time complexity of O(1) per matching computation per port, via parallelization. We then propose another algorithm called Sliding-Window QPS (SW-QPS). SW-QPS retains and enhances all benefits of SB-QPS, and reduces the batching delay to zero via a novel switching framework called sliding-window switching. In addition, SW-QPS computes matchings of much higher qualities, as measured by the resulting throughput and delay performances, than QPS-1, the state-of-the-art regular switching algorithm that builds upon the same underlying bipartite matching algorithm.

Joint Rate and Fairness Improvement Based on Adaptive Weighted Graph Matrix for Uplink SCMA With Randomly Distributed Users

Developing resource allocation algorithms for the uplink sparse code multiple access (SCMA) scheme to satisfy multiple objectives is challenging, especially where users are randomly distributed. In this paper, we aim to address this challenge by developing a joint resource allocation method as a multi-objective optimization (MO) problem to maximize the average sum rate and fairness among users as key and sub-key objectives, respectively. For this purpose, the exact analytical expressions for the average sum rate and users' individual rate are extracted based on an adaptive weighted graph matrix (AWGM). An AWGM matrix beneficially replaces the factor graph and the power allocation matrices to simplify the MO problem based on the asymmetric modified bipartite matching (AMBM) algorithm. The power allocation strategy is utilized during the optimal resource assignment process using the AMBM algorithm. After the AMBM process, we propose a low-complexity four-step algorithm to obtain the AWGM. The simulation results show that our proposed method can compromise and improve the multiple objectives' performance and guarantees a stable range of network performance at different times.

An Energy-Efficient Unselfish Spectrum Leasing Scheme for Cognitive Radio Networks.

Cooperative Communications in Cognitive Radio (CR) have been introduced as an essential and efficient technique to improve the transmission performance of primary users and offer transmission opportunities for secondary users. In a typical multiuser Cooperative Communication in CR, each primary user can choose one secondary user as a relay node. To encourage the cooperative behavior of the secondary users, primary users lease a fraction of their allocated spectrum to the relay secondary users to transmit their data packets. In this work, a novel unselfish spectrum leasing scheme in CR networks is proposed that offers an energy-efficient solution minimizing the environmental impact of our network. A network management architecture is introduced, and resource allocation is proposed as a constrained sum energy efficiency maximization problem. The optimization problem is formulated and solved using non-linear programming methods and based on a modified Kuhn-Munkres bipartite matching algorithm. System simulations demonstrate an increment in the energy efficiency of the primary users’ network compared with previously proposed algorithms.

A Modified Decomposition Algorithm for Maximum Weight Bipartite Matching and Its Experimental Evaluation

Let G be an undirected bipartite graph with positive integer weights on the edges. We refine the existing decomposition theorem originally proposed by Kao et al., for computing maximum weight bipartite matching. We apply it to design an efficient version of the decomposition algorithm to compute the weight of a maximum weight bipartite matching of G in O(|V |W /k(|V |, W /N))-time by employing an algorithm designed by Feder and Motwani as a subroutine, where |V | and N denote the number of nodes and the maximum edge weight of G, respectively and k(x, y) = log x/ log(x 2 /y). The parameter W is smaller than the total edge weight W, essentially when the largest edge weight differs by more than one from the second largest edge weight in the current working graph in any decomposition step of the algorithm. In best case W = O(|E|) where |E| be the number of edges of G and in worst case W = W, that is, |E| ≤ W ≤ W. In addition, we talk about a scaling property of the algorithm and research a better bound of the parameter W. An experimental evaluation on randomly generated data shows that the proposed improvement is significant in general.

Surface shape-based clustering for B-rep models

Matching, clustering, and retrieving 3D CAD models of mechanical components based on their shape are useful for many CAD/CAM applications such as design reuse, variant process planning and group technology. Surfaces are the prominent elements of the B-Rep (Boundary Representation) CAD model, but the current methods of similarity assessment are not centered on the surfaces and lack an accurate description of their geometric features. In order to solve the problem of retrieval and clustering of B-Rep models more efficiently, the concept of “most crucial surface” is proposed and its corresponding characteristics are studied in detail. The contribution of our approach is that surfaces are the major shape determinants of the B-Rep model, and the distribution of Carosati curvatures is the optimum shape features of surfaces. First, the surface elements are extracted from the STEP (Standard for Exchange of Product Model) files of the B-rep models, and the distribution of minimum, Gauss and Carosati surface curvatures are converted into the shape feature space by the wavelet transform, the Fourier transform, and the grouping calculation. Thus we characterize the B-Rep model as a histogram with surfaces as bins, and then compare and cluster the B-Rep models by the bipartite matching algorithm or the earth mover’s distance. The surface-based methods are evaluated with the four effectiveness indices in the clustering experiment of the NDR (National Design Reservoir) data, and the results indicated that the grouping method for the surface Carosati curvatures has a highly competent matching and clustering performance.