Ensuring data ownership and traceability of unauthorised redistribution are central to safeguarding intellectual property in shared data environments. Data fingerprinting addresses these challenges by embedding recipient-specific marks into the data, typically via content modifications. We propose NCorr-FP, a Neighbourhood-based Correlation-preserving Fingerprinting system for structured tabular data that preserves the statistical fidelity of data. The method uses local record similarity and density estimation to guide the insertion of fingerprint bits. The embedding logic is then reversed to extract the fingerprint from a potentially modified dataset. Extensive experiments confirm its effectiveness, fidelity, utility, and robustness. Results show that fingerprints are virtually imperceptible, with minute Hellinger distances and KL divergences, even at high embeddeding rates. The system also maintains high data utility for downstream predictive tasks. The method achieves 100% detection confidence under substantial data deletions and remains robust against adaptive- and collusion attacks. Satisfying all these requirements concurrently on mixed-type datasets highlights the strong applicability of NCorr-FP to real-world data settings.

Accurately and efficiently identifying anomalies within log data is crucial for maintaining the reliability, availability, and security of modern computing systems. In interconnected environments, log data often come from distributed sources such as Internet of Things (IoT) devices, industrial networks, or smart grids. Centralizing these logs for anomaly detection can be challenging due to strict confidentiality requirements, regulatory constraints, and the limited bandwidth of edge networks. Federated learning (FL) offers an alternative by enabling local model training on site, while aggregating only models instead of sensitive data, thereby preserving data confidentiality and reducing data transfer. This paper develops and evaluates a federated log-anomaly detection pipeline and analyzes its components. We adapt lightweight anomaly detection techniques in a federated setting, comparing them with deep learning (DL) methods, assessing detection capabilities, computational efficiency, memory requirements, and inference time. We explore the residual privacy risks in FL within the proposed pipeline, develop a threat model and suggest mitigation strategies. Our findings indicate that lightweight anomaly detection methods can match the effectiveness of DL techniques in the FL framework, while often providing improved computational and communication efficiency. Furthermore, these techniques show better resilience to non-IID data distributions, a typical FL challenge that can severely hinder the effectiveness of traditional machine learning models. However, the memory footprint of lightweight models varies with dataset characteristics and, in some cases, may exceed that of DL models.

Cyber Situational Awareness (CSA) is an important element in both cyber security and cyber defence to inform processes and activities on strategic, tactical, and operational level. Furthermore, CSA enables informed decision making. The ongoing digitization and interconnection of previously unconnected components and sectors equally affects the civilian and military sector. In defence, this means that the cyber domain is both a separate military domain as well as a cross-domain and connecting element for the other military domains comprising land, air, sea, and space. Therefore, CSA must support perception, comprehension, and projection of events in the cyber space for persons with different roles and expertise. This paper introduces NEWSROOM, a research initiative to improve technologies, methods, and processes specifically related to CSA in cyber defence. For this purpose, NEWSROOM aims to improve methods for attacker behavior classification, cyber threat intelligence (CTI) collection and interaction, secure information access and sharing, as well as human computer interfaces (HCI) and visualizations to provide persons with different roles and expertise with accurate and easy to comprehend mission- and situation-specific CSA. Eventually, NEWSROOM’s core objective is to enable informed and fast decision-making in stressful situations of military operations. The paper outlines the concept of NEWSROOM and explains how its components can be applied in relevant application scenarios.

Due to its high potential for analysis in clinical settings, research on the human microbiome has been flourishing for several years. As an increasing amount of data on the microbiome is gathered and stored, analysing the temporal and individual stability of microbiome readings, and the succeeding privacy risks, has gained importance. In 2015, Franzosa et al. demonstrated the feasibility of matching and linking individuals in microbiome-based datasets from the Human Microbiome Project, which could lead to re-identification of individuals, and thus poses privacy implications for microbiome study designs. Their technique is based on the construction of body site-specific metagenomic codes that maintain a certain stability over time. In this paper, we establish a distance-based technique for personal microbiome identification, which is combined with a solution for avoiding spurious, false positive matches. In a direct comparison with the approach from Franzosa et al., which assumes that information is available as microbial records, rather than at the more detailed (but less likely to be shared) nucleic acid level, our method improves upon the identification results on most of the considered datasets. Our main finding is an increase of the average percentage of true positive identifications of 30% on the widely studied microbiome of the gastrointestinal tract. While we particularly recommend our method for application on the gut microbiome, we also observed substantial identification success on other body sites. Our results demonstrate the potential of privacy threats in microbiome data gathering, storage, sharing, and analysis, and thus underline the need for solutions to protect the microbiome as personal and sensitive medical data. We also show that the method is robust to various hyper-parameter settings. Based on our observations, we further identify challenges in personal microbiome identification research, specifically, the scarcity of benchmark data and associated data analysis tasks. Based on our experience, we propose solutions for a more systematic and comparable evaluation, considering also aspects of costs entailed with applying privacy-preserving methods.

The commercial use of machine learning (ML) is spreading; at the same time, ML models are becoming more complex and more expensive to train, which makes intellectual property protection (IPP) of trained models a pressing issue. Unlike other domains that can build on a solid understanding of the threats, attacks, and defenses available to protect their IP, ML-related research in this regard is still very fragmented. This is also due to a missing unified view as well as a common taxonomy of these aspects. In this article, we systematize our findings on IPP in ML while focusing on threats and attacks identified and defenses proposed at the time of writing. We develop a comprehensive threat model for IP in ML, categorizing attacks and defenses within a unified and consolidated taxonomy, thus bridging research from both the ML and security communities.

The human microbiome is increasingly subject to extensive research, due to its relations to health, diet, exercise and illness. While ever more microbiome data is gathered and stored, recent works have demonstrated the threat of individual re-identification based on matching samples taken at different points in time, by matching metagenomic features extracted from microbiome readings. The individual and temporal stability of the microbiome varies for different body sites and is particularly pronounced for readings from the gastrointestinal tract. To meet the resulting need for privacy-protecting solutions, we adapt the well-known concept of k-anonymity and make it suitable for application to microbiome datasets. In particular, our approach for establishing k-anonymity is based on micro-aggregation.Our evaluation uses ten datasets containing samples of gut microbiomes, and analyzes the decreased privacy risk on the anonymised dataset as well as the incurred information loss. The analysis demonstrates the suitability of our approach for the protection of sensitive microbiome data.

With the ever-increasing amount of data collected, there is also an increased demand for data analysis and machine learning methods, which are consequently frequently deployed. However, many of the data collected are very sensitive and of a personal nature – thus, data confidentiality and privacy become important considerations. In the wake of this, the use of synthetic data as a privacy-preserving measure for micro-data is gaining more and more popularity, especially due to its ability to maintain a high level of data utility. Synthetic data is artificially generated by a model that has been trained on real data. This means that the observations in the synthetic data do not directly correspond to any individual in the original dataset. While there are many tools for creating synthetic data available, only a little research has focused on specifically treating sensitive attributes and generating synthetic data in a way that concentrates on protecting these selected attributes from inference attacks while keeping the data utility as high as possible. This can be achieved done by setting certain constraints when learning the model from the original data. Earlier work proposed a modification to extend the DataSynthesizer, an approach for synthetic data generation that uses Bayesian Networks to capture the underlying structures in the original data, to protect one sensitive attribute. In this paper, we investigate two different techniques for extending this approach to protect multiple attributes from inference and analyse the subsequent effects on the data utility.

Federated Learning (FL) is a method that allows multiple entities to jointly train a machine learning model using data located in various places. Unlike the conventional approach of gathering private data from distributed locations to a central place, federated learning involves solely exchanging and aggregating the machine learning models. Each party shares only a machine learning model trained locally on their private data, ensuring that the sensitive data remains within the respective silos throughout the process. However, these shared models in FL may still leak sensitive information about the training data in the form of e.g. membership disclosure. To mitigate these residual privacy risks in federated learning, one has to use additional defence techniques such as Differential Privacy (DP), which introduces noise into the training data or the model. Differential Privacy provides a mathematical definition of privacy and can be applied in machine learning via different perturbation mechanisms. This work focuses on the analysis of Differential Privacy in federated learning through (i) output perturbation of the trained machine learning models and (ii) a differentially-private form of stochastic gradient descent (DP-SGD). We consider these two approaches in various settings and analyse their performance in terms of model utility and achieved privacy. To evaluate a model’s privacy risk, we empirically measure the success rate of a membership inference attack. We observe that DP-SGD allows for a better trade-off between privacy and utility in most of the considered settings. In some settings, however, output perturbation can provide a better or similar privacy-utility trade-off and at the same time better communication and computational efficiency.

Federated learning is an important technique for training language models, which are frequently used for next-word prediction since federated learning allows utilising large quantities of real-life data without compromising the privacy of the data owners. Training a model that generalises well in this setting is a challenging task due to the inherent statistical heterogeneity of the training data, and due to the hardware limitations of private mobile devices. There are different approaches that address these issues, e.g. through model selection, different aggregation and learning strategies, and update compression. In this paper, two popular model architectures, namely Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), are evaluated in centralised and federated settings. For federated learning, the vanilla Federated Averaging algorithm and two alternatives that try to address statistical heterogeneity, namely FedProx, which uses a proximal term to restrict the divergence from the global model during local model training, and Federated Attention, which has similar aims of reducing the distance between models as well to ensure faster convergence and improve generalisation, but is performing this during the aggregation station, are evaluated for their achieved perplexity and accuracy in various settings on two datasets. Based on these results, we provide guidelines on which methods to use, depending on the scenario.

Since many types of data nowadays contain personally identifiable information about individuals, it is important to apply privacy protection techniques to mitigate disclosure risks. One approach is k-anonymity, where hiding the identity within a group of k similar entities reduces the risk of re-identification. Another risk when distributing data is the loss of control over their further re-distribution and sharing. This risk is frequently addressed by fingerprinting, a method that allows to identify the recipient of a specific copy of the data, by embedding a generally invisible mark.In this paper, we specifically implement and adapt an intrinsic fingerprint scheme that makes use of k-anonymity and the fact that multiple, differently perturbed versions of a dataset can be found that all fulfil a certain k-anonymity, and share a rather similar level of data precision. Thus, these different datasets can be seen each as a fingerprinted version of the original.One research question we address in this paper is the evaluation of the most common generalisation algorithms according to their generalisation strategy and their influence on data utility and the number of resulting release candidates, i.e. fingerprints and execution time. In addition, we investigate the properties and robustness of these fingerprints against intentional (adversarial) manipulation through attack simulations. We further provide recommendations and guidelines on how fingerprinting can be best achieved based on the results of our evaluation.

Federated learning provides the solution when multiple parties want to collaboratively train a machine learning model without directly sharing sensitive data. In Federated Learning, each party trains a machine learning model locally on its private data and sends only the models’ weights or updates (gradients) to an aggregator, which averages locally trained models into a new global model with higher effectiveness. However, the machine learning models, which have to be shared during the federated learning process, can still leak sensitive information about their training data through e.g. membership inference attacks. Differential Privacy (DP) can mitigate privacy risks in federated learning by introducing noise into machine learning models. In this work, we consider two approaches for achieving Differential Privacy in federated learning: (i) output perturbation of the trained machine learning models and (ii) a differentially-private form of stochastic gradient descent (DP-SGD). We perform an extensive analysis of these two approaches in several federated settings and compare their performance in terms of model utility and achieved privacy. We observe that DP-SGD allows for a better trade-off between privacy and utility.

Due to its high potential for analysis in clinical settings, research on the human microbiome has been flourishing for several years. As an increasing amount of data on the microbiome is gathered and stored, analysing the temporal and individual stability of microbiome readings and the ensuing privacy risks has gained importance. In 2015, Franzosa et al. demonstrated the feasibility of microbiome-based identifiability on datasets from the Human Microbiome Project, thus posing privacy implications for microbiome study designs. Their technique is based on the construction of body site-specific metagenomic codes that maintain a certain stability over time.

The use of synthetic data is a widely acknowledged privacy-preserving measure that reduces identity and attribute disclosure risks in micro-data. The idea is to learn the statistical properties of an original dataset, store this information in a model, and then use this model to generate artificial samples and build a synthetic dataset that resembles the original. One of the many different approaches of synthetization tools relies on describing the original dataset by using a Bayesian network. This method is implemented in the open-source tool DataSynthesizer and has proven particularly suitable for datasets with a small to moderate number of attributes. In this paper, we will substitute the greedy algorithm used for learning the Bayesian network by a substantially faster genetic algorithm. In addition, our goal is to protect particularly sensitive attributes by decreasing specific correlations in the synthetic data that may reveal personal information. We will thus show how to customize the network structures for specific machine learning tasks. Our experiments demonstrate that this technique allows to further decrease the disclosure risks and, hence, add to the applicability of synthetic data as technique for privacy preservation.

Personal, sensitive information contained in data sets is often discouraging the exchange and sharing of data, or even rendering it impossible. To still enable data sharing, anonymisation is a strategy often employed to avoid possible record identification o r i nference. A nonymisation s trategies are often data-type or modality dependent, as besides the actual attributes contained within a dataset, also certain other aspects might reveal information on the data subjects. For example in graph data, such as knowledge graphs, the structure within the graph, i.e. the connection between nodes, might allow to re-identify a specific p erson, e .g. b y k nowledge o f t he n umber of connections for some individuals within the dataset.Therefore, also the structure needs to undergo anonymisation to achieve privacy. In this paper, we optimise an algorithm that extended previous state of the art by considering multiple, different types of connections (relations) between nodes to achieve anonymity among each of these types. Our novel, open-source implementation scales to much larger graphs than previous work, which is important for efficiently a nonymising ever-increasing volumes of big, linked data.

Fingerprinting is a method of embedding a traceable mark into digital data to (i) verify the owner and (ii) identify the recipient of a released copy of a data set. This is crucial when releasing data to third parties, especially if it involves a fee, or if the data is of sensitive nature and further sharing and leaks should be discouraged and deterred from. A fingerprint is required to (i) be robust against modifications t o t he d ata to achieve successful ownership protection, while (ii) affecting the quality and utility of the data as little as possible.So far, literature mostly assumes attackers with rather limited capabilities who perform random modification t o t he dataset. With a certain task in mind to perform on the data, the attacker can however perform an adaptive and targeted attack that maximises its chances of removing or invalidating the fingerprint, while reducing the data utility the least. In the same line, the data owner can optimise the robustness of the scheme by anticipating a specific f ocus o f t he a ttacker a nd f ocusing t he fingerprint embedding on the most valuable parts of the data. In this paper, we, therefore, provide an in-depth discussion on threat models, targeted attacks and adaptive defences. We further demonstrate the impact of targeted attacks on classical and, in comparison, adaptive fingerprinting i n a n e mpirical manner.

With increasing data collection, also efforts to extract the underlying knowledge increase. Among these, collaborative learning efforts become more important, where multiple organisations want to jointly learn a common predictive model, e.g. to detect anomalies or learn how to improve a production process. Instead of learning only from their own data, a collaborative approach enables the participants to learn a more generalising model, also capable to predict settings not yet encountered by their own organisation, but some of the others. However, in many cases, the participants would not want to directly share and disclose their data, for regulatory reasons, or because the data constitute a business asset. Approaches such as federated learning allow to train a collaborative model without exposing the data itself. However, federated learning still requires exchanging intermediate models from each participant. Information that can be inferred from these models is thus a concern. Threats to individual data points and defences have been studied e.g. in membership inference attacks. However, we argue that in many use cases, also global properties are of interest -- not only to outsiders, but specifically also to the other participants, which might be competitors. In a production process, e.g. knowing which types of steps a company performs frequently, or obtaining information on quantities of a specific product or material a company processes, could reveal business secrets, without needing to know details of individual data points.

Adversarial machine learning and the robustness of machine learning is gaining attention, especially in image classification. Attacks based on data poisoning, with the aim to lower the integrity or availability of a model, showed high success rates, while barely reducing the classifiers accuracy - particularly against Deep Learning approaches such as Convolutional Neural Networks (CNNs). While Deep Learning has become the most prominent technique for many pattern recognition tasks, feature-extraction based systems still have their applications - and there is surprisingly little research dedicated to the vulnerability of those approaches. We address this gap and show preliminary results in evaluating poisoning attacks against feature-extraction based systems, and compare them to CNNs, on a traffic sign classification dataset. Our findings show that feature-extraction based ML systems require higher poisoning percentages to achieve similar backdoor success, and also need a consistent (static) backdoor position to work.

Digitalization of knowledge work in communication-intensive domains such as intellectual property protection poses great challenges but also opportunities to improve today’s working environments. The legal domain is strongly characterized by knowledge work, whereby, despite a common legal framework, creativity of individual experts is decisive. This knowledge-intensive work deals with a great amount of data objects, not only as a working basis, but also as a result. While experts heavily follow individual working styles, they still rely on a vast amount of administrative tasks, which are carried out by the supporting staff. These tasks are expected to be performed regularly, reliably and without errors, despite necessary adjustments to the current case and the changing legal framework. Today, knowledge work and administrative tasks are typically supported by different tools that are hardly integrated. Therefore, the tracing of continuous work processes based on exchanged data objects is a great challenge. This traceability is crucial, not only for legal security reasons, but also to enable mining and learning of applicable knowledge about processes. In this paper, we propose a bottom-up approach, which applies a continuously evolving graph of integrated data objects and tasks to model and store static and dynamic aspects of administrative as well as knowledge work, and test the approach in a real-world setting in the domain of intellectual property. We further present initial results of a novel dependency-based mining approach to learn data-dependent task sequences in the graph-based model and discuss several methods for enabling privacy-preserving sharing and mining.

The generation of synthetic data is widely considered as viable method for alleviating privacy concerns and for reducing identification and attribute disclosure risk in micro-data. The records in a synthetic dataset are artificially created and thus do not directly relate to individuals in the original data in terms of a 1-to-1 correspondence. As a result, inferences about said individuals appear to be infeasible and, simultaneously, the utility of the data may be kept at a high level. In this paper, we challenge this belief by interpreting the standard attacker model for attribute disclosure as classification problem. We show how disclosure risk measures presented in recent publications may be compared to or even be reformulated as machine learning classification models. Our overall goal is to empirically analyze attribute disclosure risk in synthetic data and to discuss its close relationship to data utility. Moreover, we improve the baseline for attribute disclosure risk from the attacker's perspective by applying variants of the RadiusNearestNeighbor and the EnsembleVote classifier.

Federated Learning has recently gained attraction as a means to analyze data without having to centralize it from initially distributed data sources. Generally, this is achieved by only exchanging and aggregating the parameters of the locally learned models. This enables better handling of sensitive data, e.g. of individuals, or business related content. Applications can further benefit from the distributed nature of the learning by using multiple computer resources, and eliminating network communication overhead. Adversarial Machine Learning in general deals with attacks on the learning process, and backdoor attacks are one specific attack that tries to break the integrity of a model by manipulating the behavior on certain inputs. Recent work has shown that despite the benefits of Federated Learning, the distributed setting also opens up new attack vectors for adversaries. In this paper, we thus specifically study this manipulation of the training process to embed a backdoor on the example of a dataset for traffic sign classification. Extending earlier work, we specifically include the setting of sequential learning, in additional to parallel averaging, and perform a broad analysis on a number of different settings.

With the recent advances and increasing activities in data mining and analysis, the protection of the privacy of individuals is crucial. Several approaches address this concern, from techniques like data anonymisation to secure, non-disclosive computation, all of which have their specific strengths and weaknesses, depending on the specific requirements. A slightly different approach is the generation of synthetic data, which tries to preserve the overall properties and characteristics of the original data without revealing information about actual individual data samples. The promise is that, for most purposes, models trained on the synthetic data instead of the real data do not show a significant loss of performance. In this paper, we give an overview on currently available approaches for synthetic data generation, and empirically evaluate the utility of the generated synthetic data by testing them on a number of supervised machine learning tasks on several publicly available datasets.

With ever increasing capacity for collecting, storing, and processing of data, there is also a high demand for intelligent data analysis methods. While there have been impressive advances in machine learning and similar domains in recent years, this also gives rise to concerns regarding the protection of personal and otherwise sensitive data, especially if it is to be analysed by third parties. Besides anonymisation, which becomes challenging with high dimensional data, one approach for privacy-preserving data mining lies in the usage of synthetic data, which comes with the promise of protecting the users' data and producing analysis results close to those achieved by using real data. In this paper, we analyse a number of different approaches for creating synthetic data, and study the utility of the created datasets for regression tasks, i.e. the prediction of a numeric value. We further investigate the similarity of real and synthetic data samples. Finally, we contribute to privacy assessments and measurements of the risk of attribute disclosure on synthetic data by extending an approach developed for categorical data.

We propose and evaluate a method to authenticate individuals by their unique hand dynamics, based on measurements from wearable sensors. Our approach utilises individual characteristics of hand movement when opening a door. We implement a sensor-fusion machine learning algorithm to classify individuals based on their hand movement and conduct a lab study with 20 participants to test the feasibility of the concept in the context of accessing physical doors as found in office buildings. Our results show that our approach yields an accuracy of 92% in classifying an individual and thus highlights the potential for behavioral hand dynamics for authentication.

Workflows have become a popular means for implementing experiments in computational sciences. They are beneficial over other forms of implementation, as they require a formalisation of the experiment process, they provide a standard set of functions to be used, and provide an abstraction of the underlying system. Thus, they facilitate understandability and repeatability of experimental research. Also, additional meta data standards such as Research Objects, which allow to add more meta-data about the research process, shall enable better reproducibility of experiments. However, as several studies have shown, merely implementing an experiment as a workflow in a workflow engine is not sufficient to achieve these goals, as still a number of challenges and pitfalls prevail. In this paper, we want to quantify how many workflow executions are easy to repeat. To this end, we automatically obtain and analyse a set of almost 1,500 workflows available in the myExperiment platform, focusing on the ones authored in the Taverna workflow language. We provide statistics on the types of processing steps used, and investigate what vulnerabilities in regards to re-execution are faced. We then try to automatically execute the workflows. Form these results, we conclude which are the most common causes for failures, and analyse how these can be countered, with existing or yet to be developed approaches.

In privacy sensitive eScience domains the disclosure of data is often not allowed or advised if it contains sensitive data about the individual. Applying data protection methods oppose interests of repeatability and reproducibility, as the data which serves as input and output for processing steps in experiments needs to be altered in order to preserve privacy. We thus discuss pre-requisites and methods for protecting the privacy, with the goal of still enabling reproducibility and allowing comparative studies to be performed. We align our discussion on with the experience from a use case in the area of health policy planning, where statistical and mathematical models are trained from health data.

The re-usability and repeatability of e-Science experiments is widely understood as a requirement of validating and reusing previous work in data-intensive domains. Experiments are, however, often complex chains of processing, involving a number of data sources, computing infrastructure, software tools, or external and third-party services, rendering repeatability a challenging task. Another important aspect of many experiments is in the social and organisational dimension - very often, knowledge on how experiments are performed is tacit and remains with the researcher, and the collaborative and distributed aspects especially of larger collaborative experiments adds to this challenge. Therefore, a number of approaches have tackled this issue from various angles -- initiatives for data sharing, code versioning and publishing as open source, the use of workflow engines to formalise the steps taken in an experiment, to ways to describe the complex environment an experiment is executed in, e.g. via Research Objects. In this paper, we present a model that has a specific focus on the technical infrastructure that is the basis of the research experiment. We demonstrate how this model can be applied to describe e-Science experiments, and align and compare it to Research Objects.

Many business and scientific processes make extensive use of service-oriented architectures, using distributed services. These are often provided by third parties and are thus not under direct control of process owners. In this paper we discuss the issues of ensuring continuous and faithful execution of processes in distributed environments, focusing specifically on Web Services. Recently, we introduced a specification of Resilient Web Services, that makes current Web Services more robust, and a framework for the monitoring of Web Services, that allows detecting anomalies. In this paper, we describe alternative implementations of the framework for monitoring of Web Services. We also present possible approaches easing the deployment of Resilient Web Services: a framework consisting of tools deployable at the Web Service operator site enabling easy transformation of a regular Web Service into a Resilient Web Service, and a registry with notifications that decorates existing Web Services with resilient methods.

In many areas multimedia technology has made its way into mainstream. In the case of digital audio this is manifested in numerous online music stores having turned into profitable businesses. The widespread user adaption of digital audio both on home computers and mobile players show the size of this market. Thus, ways to automatically process and handle the growing size of private and commercial collections become increasingly important; along goes a need to make music interpretable by computers. The most obvious representation of audio files is their sound - there are, however, more ways of describing a song, for instance its lyrics, which describe songs in terms of content words. Lyrics of music may be orthogonal to its sound, and differ greatly from other texts regarding their (rhyme) structure. Consequently, the exploitation of these properties has potential for typical music information retrieval tasks such as musical genre classification; so far, there is a lack of means to efficiently combine these modalities. In this paper, we present findings from investigating advanced lyrics features such as the frequency of certain rhyme patterns, several parts-of-speech features, and statistic features such as words per minute (WPM). We further analyse in how far a combination of these features with existing acoustic feature sets can be exploited for genre classification and provide experiments on two test collections.

The self-organising map is a popular unsupervised neural network model which has successfully been used for clustering various kinds of data. To help in understanding the influence of single variables or components on clusterings, we introduce a novel method for the visualisation of component planes for SOMs. The approach presented is based on the discretisation of the components and makes use of the well-known metro map metaphor. It depicts consistent values and their ordering across the map for discretisations of various components and their correlations in terms of directions on the map. In our approach component lines are drawn for each component of the data, allowing the combination of numerous component planes into one plot. We also propose a method to further aggregate these component lines, by grouping highly correlated variables, i.e. similar lines on the map. To show the applicability of our approach we provide experimental results for two popular machine learning data sets.