Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks148 from public data
- 431
- Credit Card Fraud Detection Using Convolutional Neural Networks313
A CNN-based fraud detection framework, to capture the intrinsic patterns of fraud behaviors learned from labeled data is proposed, on which a convolutional neural network is applied to identify a set of latent patterns for each sample.
- 216
- Graph neural networks for financial fraud detection: a review173
This review reveals that GNNs are exceptionally adept at capturing complex relational patterns and dynamics within financial networks, significantly outperforming traditional fraud detection methods.
- F-HMTC: Detecting Financial Events for Investment Decisions Based on Neural Hierarchical Multi-Label Text Classification160
A LSTM Relational Graph Convolutional Network model, which models the connection among stocks with their correlation matrix, which outperforms the baseline models and enables the movement of stocks that are not directly associated with news as well as the whole market, which is not available in most previous methods.
- Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems156
This work is the first to identify and formally define the communication redundancy issue present in current LLM-based multi-agent pipelines, and efficiently performs one-shot pruning on the spatial-temporal message-passing graph, yielding a token-economic and high-performing communication topology.
- Temporal and Heterogeneous Graph Neural Network for Financial Time Series Prediction151
This paper proposes a temporal and heterogeneous graph neural network-based (THGNN) approach to learn the dynamic relations among price movements in financial time series and deploys the proposed THGNN in a real-world quantitative algorithm trading system, where the accumulated portfolio return obtained by the method significantly outperforms other baselines.
- Semi-supervised Credit Card Fraud Detection via Attribute-Driven Graph Representation147
The proposed semi-supervised graph neural network, namely GTAN, outperforms other state-of-the-art baselines on three fraud detection datasets and model the fraud patterns through risk propagation among transactions.
- Spatio-Temporal Attention-Based Neural Network for Credit Card Fraud Detection126
The result shows that STAN performs better than other state-of-the-art baselines in both AUC and precision-recall curves and demonstrates the effectiveness of the proposed method in both detecting suspicious transactions and mining fraud patterns.
- G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks125
G-Designer is introduced, an adaptive, efficient, and robust solution for multi-agent deployment, which dynamically designs task-aware, customized communication topologies.
- Knowledge Graph-based Event Embedding Framework for Financial Quantitative Investments106
A knowledge graph-based event embedding framework for quantitative investments that first extracts structured events from raw texts, and construct the knowledge graph with the mentioned entities and relations simultaneously, and leverages a joint model to merge theknowledge graph information into the objective function of anevent embedding learning model.
- TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecasting101
TimeFilter is proposed, a GNN-based framework for adaptive and fine-grained dependency modeling that refines the learned spatial-temporal dependencies by filtering out irrelevant correlations while preserving the most critical ones in a patch-specific manner.
- 84
- Risk Assessment for Networked-guarantee Loans Using High-order Graph Attention Representation65
A high-order graph attention representation method (HGAR) to learn the embedding of guarantee networks and proves that the complexity is nearlinear with the number of edges, which could scale to large datasets.
- AgriKG: An Agricultural Knowledge Graph and Its Applications65
An agricultural knowledge graph, namely AgriKG, is proposed to automatically integrate the massive agricultural data from internet by applying the NLP and deep learning techniques, and can automatically recognize agricultural entities from unstructured text, and link them to form a knowledge graph.
- Image Cropping with Composition and Saliency Aware Aesthetic Score Map59
This paper proposes an interpretable image cropping model that can be used to localize aesthetically important regions in an image, which sheds light on the composition rules learned by the model.
- 58
- 52
- Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learning47
An innovative framework that adapts Large Language Models for time-series representation learning with a distinctive element of the patch-wise decoding layer, which departs from previous methods reliant on sequence-level decoding.
- 39
- Efficient Top-k Vulnerable Nodes Detection in Uncertain Graphs39
The proposed model can better estimate the default risks of enterprises compared to the state-of-the-art techniques and is integrated with the loan risk control system, which is deployed in the collaborated bank.
- 37
- 36
- A Dynamic Default Prediction Framework for Networked-guarantee Loans35
A dynamic default prediction framework (DDPF) is proposed, which preserves temporal network structures and loan behavior sequences in an end-to-end model and design a gated recursive and attention mechanism to integrate both the loan behavior and network information.
- Contagious Chain Risk Rating for Networked-guarantee Loans34
A novel approach to rate the risk of contagion chains in the bank industry with the deep neural network is proposed and is significantly better than the state-of-the-art baselines on the dataset from a major financial institution in Asia.
- 34
- 32
- A comprehensive and practical benchmark for financial time series forecasting31
The proposed Financial Time Series Benchmark (FinTSB) is a comprehensive and practical benchmark for financial time series forecasting (FinTSF), and provides researchers with a novel and comprehensive platform for improving and evaluating FinTSF methods.
- MCI-GRU: Stock prediction model based on multi-head cross-attention and improved GRU31
A stock prediction model, MCI-GRU, based on a multi-head cross-attention mechanism and an improved GRU, that outperforms SOTA techniques across multiple metrics is proposed and its successful application in real-world fund management operations confirms its effectiveness and practicality.
- Pre-trained Online Contrastive Learning for Insurance Fraud Detection30
This paper proposes an innovative online learning method for medical insurance fraud detection, named POCL, which combines contrastive learning pre-training with online updating strategies and has significant advantages in accuracy compared to the state-of-the-art baseline methods, while also exhibiting lower running time and space consumption.
- LSR-IGRU: Stock Trend Prediction Based on Long Short-Term Relationships and Improved GRU29
This paper proposes a stock price trend prediction model named LSR-IGRU, which is based on long short-term stock relationships and an improved GRU input, and validates the superiority of the proposed model over the current state-of-the-art baseline models.
- Risk Guarantee Prediction in Networked-Loans29
An attention-based graph neural network to encode the edges that preserve the financial status as well as network structures in dynamic loan networks and shows that DGANN could significantly improve the risk prediction accuracy compared with state-of-the-art baselines.
- RA-CFGPT: Chinese financial assistant with retrieval-augmented large language model27
The RA-CFGPT system, designed for Chinese financial tasks such as question answering, document analysis, investment advising, and risk assessment, combines a hybrid knowledge base, a fine-tuned Chinese FinLLM, an information organizer, and response checkers to ensure outputs are accurate, compliant, and highlight potential risks.
- CFBenchmark: Chinese Financial Assistant Benchmark for Large Language Model27
The experimental results indicate that while some LLMs show outstanding performance in specific tasks, overall, there is still significant room for improvement in basic tasks of financial text processing with existing models.
- 25
- Learning From Web Data With Self-Organizing Memory Module25
A novel method, capable of handling these two types of noises together, without the supervision of clean images in the training stage, is proposed, under the framework of multi-instance learning.
- iConViz: Interactive Visual Exploration of the Default Contagion Risk of Networked-Guarantee Loans25
The default contagion risk was identified, a pivotal issue in developing preventive measures, and iConViz, an interactive visual analysis tool that facilitates the closed-loop analysis process was established to quantify the infectious consequences of guarantee chains in this type of network.
- 24
- 24
- General graph generators: experiments, analyses, and improvements24
An overview of the existing general graph generators, including recently emerged deep learning-based approaches, and a new method that can achieve a good trade-off between simulation quality and efficiency are proposed.
- CFGPT: Chinese Financial Assistant with Large Language Model23
A Chinese Financial Generative Pre-trained Transformer framework, named CFGPT, which includes a dataset for pre-training and supervised fine-tuning, a financial LLM ~(CFLLM) to adeptly manage financial texts, and a deployment framework~(CFAPP) designed to navigate real-world financial applications.
- 22
- Efficient Continuous Space Policy Optimization for High-frequency Trading22
This is the first work that solves the high-frequency portfolio optimization problem by devising an efficient continuous space policy optimization algorithm in the DRL framework and shows that the proposed DRPO significantly outperforms state-of-the-art benchmark methods.
- Fighting against Organized Fraudsters Using Risk Diffusion-based Parallel Graph Neural Network22
RDPGL, a novel Risk Diffusion-based Parallel Graph Learning approach, to fighting against medical insurance criminal gangs is proposed, which could detect more organized fraud claims with relatively high precision compared with state-of-the-art baselines.
- 21
- Modeling Similarities Among Multi-Dimensional Financial Time Series21
Experiments on stocks data of all constituent firms of the S&P 500 demonstrate the superior performance of the proposed tensor-based framework when compared with some state-of-the-art methods.
- Grad: Guided Relation Diffusion Generation for Graph Augmentation in Graph Fraud Detection20
A relation diffusion-based graph augmentation model Grad is proposed, which leverages a supervised graph contrastive learning module to enhance the fraud-benign difference and employs a guided relation diffusion generator to generate auxiliary homophilic relations from scratch.
- 20
- Effective High-order Graph Representation Learning for Credit Card Fraud Detection18
A novel High-order Graph Representation Learning model (HOGRL) is proposed to avoid incorporating excessive noise during the multi-layer aggregation process and to identify fraudsters' multi-hop indirect transactions via multi-layer feature learning.
- 18
- Two Heads Are Better Than One: Boosting Graph Sparse Training via Semantic and Topological Awareness16
This work proposes a new research line and concept termed Graph Sparse Training (GST), which dynamically manipulates sparsity at the data level, and introduces the Equilibria Sparsification Principle to guide this process.
- Span-reachability querying in large temporal graphs16
This paper defines a new reachability model, called span-reachability, designed to relax the time order dependency and identify the relationship between entities in a given time period, and proposes an index-based method to answer span- reachability queries.
- FinMamba: Market-Aware Graph Enhanced Multi-Level Mamba for Stock Movement Prediction14
The proposed FinMamba, a Mamba-GNN-based framework for market-aware and multi-level hybrid stock movement prediction, devise a dynamic graph to learn the changing representations of inter-stock relationships by integrating a pruning module that adapts to market trends.
- Efficient Dynamic Attributed Graph Generation13
VDAG is introduced, a novel variational recurrent framework for efficient dynamic attributed graph generation that avoids the time-consuming path sampling and merging process in existing random walk-based methods, significantly reducing the synthesis time.
- 13
- Preventing Attacks in Interbank Credit Rating with Selective-aware Graph Neural Network12
A novel selective-aware graph neural network model (SA-GNN) for defense the Interbank credit rating attacks and optimized the model with weighed penalization on the objection function so that the model could differentiate the attackers.
- Fusing Global Domain Information and Local Semantic Information to Classify Financial Documents12
This paper implements a document classification framework, named GraphSEAT, to classify financial documents for a leading financial information service provider in China, and performs extensive experiments on the real-world financial news dataset and three public datasets to evaluate the performance.
- 11
- Adversarial Multi-task Learning for Efficient Chinese Named Entity Recognition11
This article learns commonly shared boundary information of entities from multiple kinds of tasks and combines the learned boundary information with the semantic information to improve entity recognition with multi-task learning and proposes a compression method based on improved clustering to accelerate the proposed model.
- 10
- 10
- 10
- GDeR: Safeguarding Efficiency, Balancing, and Robustness via Prototypical Graph Pruning9
A novel dynamic soft-pruning method, GDeR, designed to update the training ``basket'' during the process using trainable prototypes, which achieves the goal of Graph Training Debugging.
- Optimal Action Space Search9
This paper proposes an end-to-end DRL method that explores solutions on the whole graph via a probabilistic dynamic programming algorithm and can generate stable trading strategies for both high-frequency and low-frequency trading, significantly outperforming the baseline DRL methods on annualized return and Sharpe ratio.
- Critical Firms Prediction for Stemming Contagion Risk in Networked-Loans through Graph-Based Deep Reinforcement Learning8
This work proposes a novel approach to predict critical firms for stemming contagion risk in the bank industry with deep reinforcement learning integrated with high-order graph message-passing networks and demonstrates that it outperforms the state-of-the-art baselines significantly on the dataset from a large commercial bank.
- Visual analytics for loan guarantee network risk management.8
This is the first visual analytics tool to explore the guarantee network risks in a systematic manner and can be useful for the government and bank to monitor the default spread status and the temporal diffusion path analysis.
- User behavior detection framework based on NBP for energy efficiency8
A user behavior detection framework based on non-parametric belief propagation is proposed to discover the anomalous events from the dataset of contextual user behaviors, as in the case of anomalies that can make a surge of energy flow and mess the energy management schedule.
- 7
- 7
- 7
- Exploiting Motion Information from Unlabeled Videos for Static Image Action Recognition7
This paper integrates the above two strategies in a unified framework, which consists of Visual Representation Enhancement (VRE) module and Motion Representation Augmentation (MRA) module, and demonstrates the superiority of the framework based on four benchmark human action datasets with limited labeled data.
- A hybrid approach for risk assessment of loan guarantee network7
A hybrid feature representation is proposed, which is feeded into a gradient boosting model for credit risk assessment of guarantee network, and shows promising prediction performance on real-world data, which can be useful to both regulators and stakeholders.
- CFBenchmark-MM: Chinese Financial Assistant Benchmark for Multimodal Large Language Model6
CFBenchmark-MM is introduced, a Chinese multimodal financial benchmark with over 9,000 image-question pairs featuring tables, histogram charts, line charts, pie charts, pie charts, and structural diagrams, and a staged evaluation system to assess MLLMs in handling multimodal information by providing different visual content step by step.
- Can LLMs Alleviate Catastrophic Forgetting in Graph Continual Learning? A Systematic Study6
This work attempts to answer whether large language models (LLMs) can mitigate catastrophic forgetting in Graph Continual Learning (GCL), and proposes a simple-yet-effective method, Simple Graph Continual Learning (SimGCL), that surpasses the previous state-of-the-art GNN-based baseline by around 20% under the rehearsal-free constraint.
- 5
- 5
- 5
- 5
- 5
- 5
- BHONEM: Binary High-Order Network Embedding Methods for Networked-Guarantee Loans5
A binary higher-order network embedding method to learn the low-dimensional representations of a guarantee network and shows that this method outperforms other start-of-the-art algorithms for both classification accuracy and robustness, especially in the guarantee network.
- Multi-Granularity Augmented Graph Learning for Spoofing Transaction Detection4
This work proposes a multi-granularity augmented graph learning method that differentially captures fraud signals at local, group, and global levels, using multi-hop differential aggregation and community-augmented strategy to capture information from local to global perspectives.
- Dual Pairwise Pre-training and Prompt-tuning with Aligned Prototypes for Interbank Credit Rating4
A Dual Pairwise pre-training and prompt Tuning framework with Aligned Prototypes (DPTAP) for interbank credit rating, which enables dynamic credit updates and demonstrates the superiority of the proposed framework over various competitive models in early warning and risk contagion forecasting.
- 4
- Dynamic Graph-based Deep Reinforcement Learning with Long and Short-term Relation Modeling for Portfolio Optimization4
A Dynamic Graph-based Deep Reinforcement Learning (DGDRL) for optimal portfolio decisions is proposed by devising two mechanisms for naturally modeling the financial market by utilizing the static and dynamic graphs to represent the long and short-term relations.
- Generation is better than Modification: Combating High Class Homophily Variance in Graph Anomaly Detection4
A novel GNN model named HedGe, which improves the robustness under the novel Heterophily Attack with increased class homophily variance on other graph classification tasks, and improves the robustness under the novel Heterophily Attack with increased class homophily variance on other graph classification tasks.
- Beyond Knowledge to Agency: Evaluating Expertise, Autonomy, and Integrity in Finance with CNFinBench3
The Harmful Instruction Compliance Score (HICS), a multi-dimensional safety metric that integrates risk-type-specific deductions, multi-turn consistency tracking, and severity-adjusted penalty scaling based on fine-grained violation triggers, is introduced.
- Efficient Approximate Nearest Neighbor Search under Multi-Attribute Range Filter3
KHI is proposed, an index for multi-attribute RFANNS that combines an attribute-space partitioning tree with HNSW graphs attached to tree nodes that consistently achieves high query throughput while maintaining high recall.
- Neighbor-enhanced Graph Pre-training and Prompt Learning Framework for Fraud Detection3
GPCF first incorporates a meta-learning-based strategy within neighbor-enhanced contrastive learning to pre-train the GNN model across diverse transaction networks, and aligns fraud detection tasks with the well-pre-trained model by simply fine-tuning the prompts.
- FinLMM-R1: Enhancing Financial Reasoning in LMM through Scalable Data and Reward Design3
An integrated framework, FinLMM-R1, combining an automated and scalable pipeline for data construction with enhanced training strategies to improve the multimodal reasoning of LMM is proposed, significantly improve answer accuracy and reasoning depth over existing reasoning LMMs in both general and financial multimodal contexts.
- 3
- 3
- Enhancing Portfolio Optimization via Heuristic-Guided Inverse Reinforcement Learning with Multi-Objective Reward and Graph-based Policy Learning3
This work proposes a novel heuristic-guided inverse reinforcement learning framework for portfolio optimization that provides an interpretable expert strategy generation mechanism that takes into account sector diversification and correlation constraints and utilizes heterogeneous graph policy learning with hierarchical attention mechanisms to explicitly model inter-stock relationships.
- Improving Anomaly Detection in Smart Grid Big Data with Knowledge Graphs3
Aiming at the power database table processing process that will cause errors, this paper constructs the knowledge graph and uses the Neo4j diagram database for storage, so as to visualize the data processing process, clarify the dataprocessing relationship, and facilitate the tracking and traceability of the data.
- VAP: A Visual Analysis Tool for Energy Consumption Spatio-temporal Pattern Discovery3
A visual analysis tool, VAP, that allows users to explore the dynamics of urban energy use at different spatial and temporal scales and focuses on analytical thinking, user interactions and answering business questions by examining different visual analysis views.
- ProtoNet: Learning from Web Data with Memory3
The novel ProtoNet is proposed, which is capable of handling these two types of noises together, without the supervision of clean images in the training stage, and can be easily integrated into arbitrary CNN model.
- Role Perceptual Augmented Temporal Graph Network for Related-party Transaction Detection2
An innovative Role Perceptual Augmented Temporal Graph Network (RPATGN) for proactive RPT detection that integrates a variational graph recurrent neural network, enhanced by temporal contextual attention, to explicitly model the dynamics of the roles and the overall network evolution.
- 2
- 2
- Generative Dynamic Graph Representation Learning for Conspiracy Spoofing Detection2
A novel framework called the Generative Dynamic Graph Model (GDGM), which models dynamic trading behaviors and the relationships among nodes to learn representations for conspiracy spoofing detection, is proposed, which outperforms state-of-the-art models in detection accuracy.
- 2
- Efficient Learning-Based Graph Simulation for Temporal Graphs2
This paper proposes an efficient learning-based approach to generate graph snapshots, namely temporal graph autoencoder (TGAE), an attention-based graph encoder to encode temporal and structural characteristics on sampled ego-graphs, and an ego-graph decoder that can achieve a good trade-off between simulation quality and efficiency in temporal graph generation.
- 2
- Data-driven power load forecasting with neural attention mechanism2
A data-driven model for power load prediction utilizing the attention mechanism, which surpasses baseline models across all evaluation metrics, offering dependable predictive support for the stable functioning of power systems.
- 2
- Learning Invariant Representations for New Product Sales Forecasting via Multi-Granularity Adversarial Learning2
A Multi-granularity AdversaRial Learning framework (MARL) is proposed to leverage knowledge from old products and improve the quality of invariant representations for more accurate sales predictions, demonstrating that the method is more effective than the existing state-of-the-art baselines for new product sales forecasting.
- 2
- 2
- Learning from Web Data with Memory Module2
A novel method, capable of handling these two types of noises together, without the supervision of clean images in the training stage, is proposed, under the framework of multi-instance learning.
- 1
- 1
- 1
- 1
- Cross-contextual Sequential Optimization via Deep Reinforcement Learning for Algorithmic Trading1
This work proposes a DRL-based cross-contextual sequential optimization (CCSO) method for algorithmic trading and designs a dynamic programming based learning method to address the challenge of complex network updates in reinforcement learning, leading to considerable enhancement in learning stability and efficiency.
- 1
- 1
- 1
- 1
- Leveraging Domain Information to Classify Financial Documents via Unsupervised Graph Momentum Contrast1
This paper proposes a graph representation learning model, namely G-MoCo, based on unsupervised graph momentum contrast, that could extract latent features from massive unlabeled raw data, and then further use the learned representations for document classification.
- 1
- Learning Dynamic Models of Gaze Point Mapping from Eye Movement1
A dynamic learning model is constructed by pupil coordinate data collected, denoising them by density-based cluster algorithm, and learning them with a mapping mechanism for gaze point mapping based on capturing eye movement from front camera of mobile devices.
- A cluster information navigate method by gaze tracking1
A new method which displays cluster information in a hierarchy pattern and interact with them by eyes' movement captured by the front camera of mobile devices is presented, providing users a new interacting method to navigate and select data quickly by eyes without any additional equipment.
- On querying minimum spanning tree in temporal graphs–
This paper investigates the MST problem in the context of temporal graphs given an undirected weighted temporal graph and proposes several index-based query algorithms with provable complexity bounds that significantly outperform the baseline online algorithm.
- –
- –
- –
- Credit and Power Co-evolution Modeling with Dynamic Graph Learning–
CPDGL is proposed, a credit-electricity co-evolution framework based on dynamic graph learning, which simultaneously performs power forecasting and credit risk assessment within a unified multi-task system and achieves state-of-the-art performance in both power forecasting and credit assessment tasks.
- Deep Reinforcement Learning Enhanced Semi-supervised Graph Neural Network for Credit Card Fraud Detection–
This work proposes DRESS, a Deep Reinforcement Learning (DRL) Enhanced Semi-supervised GNN framework that employs a DRL agent to automatically capture and enhance feature-level risks, fusing them with graph-based structural risks and propagating via a gated temporal attention network for final prediction.
- Empowering Credit Risk Detection in Weixin Pay with Billion-Scale Deep Graph Learning–
A risk-aware overlapping subgraph learning framework for large-scale credit risk detection, and a cross-subgraph consistency alignment mechanism that harmonizes the local representations into a globally consistent latent space.
- Conspiracy Spoofing Detection via Structure-Augmented Generative Graph Model–
This paper introduces the Structure-Augmented Generative Graph Model (SAG^2M) to detect conspiracy spoofing through substructure frequency-augmented detection, and demonstrates its superior effectiveness in identifying complex fraudulent behaviors.
- SoftHist: Teaching Graph Injection Attackers to Camouflage with Memory–
SoftHist, a novel gradient-free reinforcement learning framework for black-box graph injection attacks, is introduced, a novel gradient-free reinforcement learning framework for black-box graph injection attacks that combines attack effectiveness, stealthiness, and robustness against defensive measures.
- –
- Robust Graph Learning on the Web: Challenges, Methods, and Applications–
This tutorial presents a structured taxonomy of the principal robustness threats specific to web contexts and categorizes current robust graph learning approaches, spanning data-level preprocessing to model-level adaptation and generalization, and discusses representative models in detail.
- –
- AdaPatch: Adaptive Patch-Level Modeling for Non-Stationary Time Series Forecasting–
AdaPatch is introduced, a novel forecasting model specifically designed to tackle non-stationary multivariate time series and proposes a patch-based predictive decoding strategy that leverages the decoder from the reconstruction branch to replace conventional point-wise forecasting with a more structured patch-level prediction mechanism.
- –
- –
- PCAN: A Pandemic-Compatible Attentive Neural Network for Retail Sales Forecasting–
A Pandemic-Compatible Attentive Network (PCAN) is proposed for retail sales forecasting by designing a fluctuation attention mechanism based on association discrepancy in the time series and introducing a novel rectification decoding strategy to indicate fluctuation points in prediction.
- Large Language Models for Time Series Analysis: Methodologies, Applications, and Emerging Challenges–
- –
- –
- –
- –
- –
- –
- –
- VulnDS: Top-k Vulnerable SME Detection System in Networked-Loans.–
VulnDS: a top-k vulnerable SME detection system for large-scale financial networks, which is deployed in the authors' collaborated bank and can achieve up to 100x speedup ratio compared with baseline methods.
- A new collaborative sketching method on web browser–
This work presents a new tool to provide high performance, low latency, and rich element collaborative sketching within browser, exploiting XMPP for instance message; web sockets for low latency sketching and web workers to offload heavy calculated sketching items to a new thread.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.