Stock Price Manipulation Detection using Generative Adversarial Networks

Leangarun, Teema; Tangamchit, Poj; Thajchayapong, Suttipong

doi:10.1109/ssci.2018.8628777

Cited by 44 publications

(30 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are two main forms of manipulations used by attackers called pump and dump and spoof trading. The main goal of pump and dump trading is to increase a stock value and then sell it once it has been increased to obtain the maximum profit possible [1]. The pump and dump manipulations have two main stages; the pumping stage, where the manipulator falsely raises the stock price and the dumping stage.…”

Section: Stock Market Fraudmentioning

confidence: 99%

“…Deep learning approaches to this problem were introduced in [1] [7] to find an innovative solution to detect stock market manipulations. A deep general adversarial network (GAN) approach was first introduced in [1] and it showed a promise in detecting pump and dump manipulations. However, the solution was limited only to detect this one type of manipulation and had a drop-in accuracy when executed on unseen data.…”

Section: Stock Market Fraudmentioning

confidence: 99%

“…There are few exposed real-world manipulations cases, and most data are partially observable (data with missing information such as the buyer or seller), which can make validation difficult. Most artificial intelligence or machine learning applications in the financial stock market domain aim to predict the value of a stock to execute a buy or a sell [1] [2]. The use of these techniques has not been explored thoroughly in detecting anomalies in financial stock data.…”

Section: Introductionmentioning

confidence: 99%

“…Current approaches in detecting anomalies in stock market data that use supervised learning [3]- [8] suffer from reliance on a labeled dataset, which cannot be obtained in many scenarios. For example, in [1], the market authority does not disclose the buyers and sellers of each exchange. Besides, there is a limited amount of stock manipulations that get exposed to the public.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Combining Artificial Immune System and Clustering Analysis: A Stock Market Anomaly Detection Model

Close¹,

Kashef²

2020

JILSA

View full text Add to dashboard Cite

Artificial intelligence research in the stock market sector has been heavily geared towards stock price prediction rather than stock price manipulation. As online trading systems have increased the amount of high volume and real-time data transactions, the stock market has increased vulnerability to attacks. This paper aims to detect these attacks based on normal trade behavior using an Artificial Immune System (AIS) approach combined with one of four clustering algorithms. The AIS approach is inspired by its proven ability to handle time-series data and its ability to detect abnormal behavior while only being trained on regular trade behavior. These two main points are essential as the models need to adapt over time to adjust to normal trade behavior as it evolves, and due to confidentiality and data restrictions, real-world manipulations are not available for training. This paper discovers a competitive alternative to the leading approach and investigates the effects of combining AIS with clustering algorithms; Kernel Density Estimation, Self-Organized Maps, Density-Based Spatial Clustering of Applications with Noise and Spectral clustering. The best performing solution achieves leading performance using common clustering metrics, including Area Under the Curve, False Alarm Rate, False Negative Rate, and Computation Time.

show abstract

Section: Stock Market Fraudmentioning

confidence: 99%

Section: Stock Market Fraudmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Combining Artificial Immune System and Clustering Analysis: A Stock Market Anomaly Detection Model

Close¹,

Kashef²

2020

JILSA

View full text Add to dashboard Cite

show abstract

“…Although, significant detection results were claimed using this model, the model is provided with the number of decomposed components from GMM which is misleading as calling any number of components as normal and the rest abnormal cannot be justified for all the feature sets without a significant criterion. Leangurun et al [20] proposed the use of generative adversarial networks for the detection of pump-and-dump manipulation scheme and achieved 68.1% detection accuracy. The authors focused their work on Thailand stock market and trained their model using LSTM as the base structure upon normal data and later tested it using a test data having both normal and abnormal trades/transactions.…”

Section: Related Workmentioning

confidence: 99%

Stock Price Manipulation Detection based on Autoencoder Learning of Stock Trades Affinity

Rizvi

Belatreche

Bouridane

et al. 2020

2020 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Stock price manipulation, a major problem in capital markets surveillance, uses illegitimate means to influence the price of traded stocks in order to reap illicit profit. Most of the existing attempts to detect such manipulations have either relied upon annotated trading data, using supervised methods, or have been restricted to detecting a specific manipulation scheme. There have been a few unsupervised algorithms focusing on general detection yet none of them explored the innate affinity among the stock trades, be it normal or manipulative. This paper proposes a fully unsupervised model based on the idea of learning the relationship among stock prices in the form of an affinity matrix. The proposed affinity matrix based features are used to train an under-fitting autoencoder in order to learn an efficient representation of the normal stock prices. A kernel density estimate of the normal trading data is used as the reconstruction error of the autoencoder. During the detection phase, the normal dataset has been injected with synthetic manipulative trades. A kernel density estimation based clustering technique is then used to detect manipulative trades based on their autoencoder representation. The proposed approach is validated on benchmark stock price data from the LOBSTER project and the obtained results show dramatic improvements in the detection performance over existing price manipulation detection techniques.

show abstract

Identification of Stock Market Manipulation with Deep Learning

Tallboys

Zhu

Rajasegarar

2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Anomaly detection is a common and critical data mining task, it seeks to identify observations that differ significantly from others. Anomalies may indicate rare but significant events that require action. Market manipulation is an activity that undermines stock markets worldwide. This paper shares five large real-world, labelled data sets of anomalous stock market data where market manipulation is alleged to have occurred. Cutting edge deep learning techniques are then shown to successfully detect the anomalous periods. An LSTM based method with dynamic thresholding is particularly promising in this domain as it was able to identify contextual local anomalies in the data quickly, taking seconds to score two years of trading data for each stock, which can often be a challenge for deep learning approaches.

show abstract

Stock Price Manipulation Detection using Generative Adversarial Networks

Cited by 44 publications

References 24 publications

Combining Artificial Immune System and Clustering Analysis: A Stock Market Anomaly Detection Model

Combining Artificial Immune System and Clustering Analysis: A Stock Market Anomaly Detection Model

Stock Price Manipulation Detection based on Autoencoder Learning of Stock Trades Affinity

Identification of Stock Market Manipulation with Deep Learning

Contact Info

Product

Resources

About