1.
Risch, J., Krestel, R.: Bagging BERT Models for Robust Aggression Identification. Proceedings of the Workshop on Trolling, Aggression and Cyberbullying (TRAC@LREC). pp. 55–61. European Language Resources Association (ELRA) (2020).
Modern transformer-based models with hundreds of millions of parameters, such as BERT, achieve impressive results at text classification tasks. This also holds for aggression identification and offensive language detection, where deep learning approaches consistently outperform less complex models, such as decision trees. While the complex models fit training data well (low bias), they also come with an unwanted high variance. Especially when fine-tuning them on small datasets, the classification performance varies significantly for slightly different training data. To overcome the high variance and provide more robust predictions, we propose an ensemble of multiple fine-tuned BERT models based on bootstrap aggregating (bagging). In this paper, we describe such an ensemble system and present our submission to the shared tasks on aggression identification 2020 (team name: Julian). Our submission is the best-performing system for five out of six subtasks. For example, we achieve a weighted F1-score of 80.3% for task A on the test dataset of English social media posts. In our experiments, we compare different model configurations and vary the number of models used in the ensemble. We find that the F1-score drastically increases when ensembling up to 15 models, but the returns diminish for more models.
2.
Risch, J., Künstler, V., Krestel, R.: HyCoNN: Hybrid Cooperative Neural Networks for Personalized News Discussion Recommendation. Proceedings of the International Joint Conferences on Web Intelligence and Intelligent Agent Technologies (WI-IAT) (2020).
Many online news platforms provide comment sections for reader discussions below articles. While users of these platforms often read comments, only a minority of them regularly write comments. To encourage and foster more frequent engagement, we study the task of personalized recommendation of reader discussions to users. We present a neural network model that jointly learns embeddings for users and comments encoding general properties. Based on explicit and implicit user feedback, we sample relevant and irrelevant reader discussions to build a representative training dataset. We compare to several baselines and state-of-the-art approaches in an evaluation on two datasets from The Guardian and Daily Mail. Experimental results show that the recommendations of our approach are superior in terms of precision and recall. Further, the learned user embeddings are of general applicability because they preserve the similarity of users who share interests in similar topics.
3.
Jain, N., Bartz, C., Bredow, T., Metzenthin, E., Otholt, J., Krestel, R.: Semantic Analysis of Cultural Heritage Data: Aligning Paintings and Descriptions in Art-Historic Collections. International Workshop on Fine Art Pattern Extraction and Recognition in conjunction with the 25th International Conference on Pattern Recognition (ICPR 2020) (2020).
Art-historic documents often contain multimodal data in terms of images of artworks and metadata, descriptions, or interpretations thereof. Most research efforts have focused either on image analysis or text analysis independently since the associations between the two modes are usually lost during digitization. In this work, we focus on the task of alignment of images and textual descriptions in art-historic digital collections. To this end, we reproduce an existing approach that learns alignments in a semi-supervised fashion. We identify several challenges while automatically aligning images and texts, specifically for the cultural heritage domain, which limit the scalability of previous works. To improve the performance of alignment, we introduce various enhancements to extend the existing approach that show promising results.
4.
Lazaridou, K., Löser, A., Mestre, M., Naumann, F.: Discovering Biased News Articles Leveraging Multiple Human Annotations. Proceedings of the Conference on Language Resources and Evaluation (LREC). pp. 1268–1277 (2020).
Unbiased and fair reporting is an integral part of ethical journalism. Yet, political propaganda and one-sided views can be found in the news and can cause distrust in media. Both accidental and deliberate political bias affect the readers and shape their views. We contribute to a trustworthy media ecosystem by automatically identifying politically biased news articles. We introduce novel corpora annotated by two communities, i.e., domain experts and crowd workers, and we also consider automatic article labels inferred by the newspapers’ ideologies. Our goal is to compare domain experts to crowd workers and also to prove that media bias can be detected automatically. We classify news articles with a neural network and we also improve our performance in a self-supervised manner.
5.
Jain, N.: Domain-Specific Knowledge Graph Construction for Semantic Analysis. Extended Semantic Web Conference (ESWC 2020) Ph.D. Symposium (2020).
Knowledge graphs are widely used for systematic representation of real world data. They serve as a backbone for a number of applications such as search, questions answering and recommendations. Large scale, general purpose knowledge graphs, having millions of facts, have been constructed through automated techniques from publicly available datasets such as Wikipedia. However, these knowledge graphs are typically incomplete and often fail to correctly capture the semantics of the data. This holds true particularly for domain-specific data, where the generic techniques for automated knowledge graph creation often fail due to novel challenges, such as lack of training data, semantic ambiguities and absence of representative ontologies. In this thesis, we focus on automated knowledge graph constriction for the cultural heritage domain. We investigate the research challenges encountered during the creation of an ontology and a knowledge graph from digitized collections of cultural heritage data based on machine learning approaches. We identify the specific research problems for this task and present our methodology and approach for a solution along with preliminary results.
6.
Risch, J., Garda, S., Krestel, R.: Hierarchical Document Classification as a Sequence Generation Task. Proceedings of the Joint Conference on Digital Libraries (JCDL). pp. 147–155 (2020).
Hierarchical classification schemes are an effective and natural way to organize large document collections. However, complex schemes make the manual classification time-consuming and require domain experts. Current machine learning approaches for hierarchical classification do not exploit all the information contained in the hierarchical schemes. During training, they do not make full use of the inherent parent-child relation of classes. For example, they neglect to tailor document representations, such as embeddings, to each individual hierarchy level. Our model overcomes these problems by addressing hierarchical classification as a sequence generation task. To this end, our neural network transforms a sequence of input words into a sequence of labels, which represents a path through a tree-structured hierarchy scheme. The evaluation uses a patent corpus, which exhibits a complex class hierarchy scheme and high-quality annotations from domain experts and comprises millions of documents. We re-implemented five models from related work and show that our basic model achieves competitive results in comparison with the best approach. A variation of our model that uses the recent Transformer architecture outperforms the other approaches. The error analysis reveals that the encoder of our model has the strongest influence on its classification performance.
7.
Risch, J., Ruff, R., Krestel, R.: Offensive Language Detection Explained. Proceedings of the Workshop on Trolling, Aggression and Cyberbullying (TRAC@LREC). pp. 137–143. European Language Resources Association (ELRA) (2020).
Many online discussion platforms use a content moderation process, where human moderators check user comments for offensive language and other rule violations. It is the moderator's decision which comments to remove from the platform because of violations and which ones to keep. Research so far focused on automating this decision process in the form of supervised machine learning for a classification task. However, even with machine-learned models achieving better classification accuracy than human experts in some scenarios, there is still a reason why human moderators are preferred. In contrast to black-box models, such as neural networks, humans can give explanations for their decision to remove a comment. For example, they can point out which phrase in the comment is offensive or what subtype of offensiveness applies. In this paper, we analyze and compare four attribution-based explanation methods for different offensive language classifiers: an interpretable machine learning model (naive Bayes), a model-agnostic explanation method (LIME), a model-based explanation method (LRP), and a self-explanatory model (LSTM with an attention mechanism). We evaluate these approaches with regard to their explanatory power and their ability to point out which words are most relevant for a classifier's decision. We find that the more complex models achieve better classification accuracy while also providing better explanations than the simpler models.
8.
Pena, E.H.M., Filho, E.R.L., de Almeida, E.C., Naumann, F.: Efficient Detection of Data Dependency Violations. Proceedings of the International Conference on Information and Knowledge Management (CIKM) (2020).
Research on data dependencies has experienced a revival as dependency violations can reveal errors in data. Several data cleaning systems use a DBMS to detect such violations. While DBMSs are efficient for some kinds of data dependencies (e.g., unique constraints), it is likely to fall short of satisfactory performance for more complex ones, such as order dependencies. We present a novel system to efficiently detect violations of denial constraints (DCs), a well-known formalism that generalizes many kinds of data dependencies. We describe its execution model, which operates on a compressed block of tuples at-a-time, and we present various algorithms that take advantage of the predicate form in the DCs to provide effective code patterns. Our experimental evaluation includes comparisons with DBMS-based and DC-specific approaches, real-world and synthetic data, and various kinds of DCs. It shows that our system is up to three orders-of-magnitude faster than the other solutions, especially for datasets with a large number of tuples and DCs that identify a large number of violations.
9.
Risch, J., Krestel, R.: A Dataset of Journalists’ Interactions with Their Readership: When Should Article Authors Reply to Reader Comments?. Proceedings of the International Conference on Information and Knowledge Management (CIKM). pp. 3117–3124. ACM (2020).
The comment sections of online news platforms are an important space to indulge in political conversations andto discuss opinions. Although primarily meant as forums where readers discuss amongst each other, they can also spark a dialog with the journalists who authored the article. A small but important fraction of comments address the journalists directly, e.g., with questions, recommendations for future topics, thanks and appreciation, or article corrections. However, the sheer number of comments makes it infeasible for journalists to follow discussions around their articles in extenso. A better understanding of this data could support journalists in gaining insights into their audience and fostering engaging and respectful discussions. To this end, we present a dataset of dialogs in which journalists of The Guardian replied to reader comments and identify the reasons why. Based on this data, we formulate the novel task of recommending reader comments to journalists that are worth reading or replying to, i.e., ranking comments in such a way that the top comments are most likely to require the journalists' reaction. As a baseline, we trained a neural network model with the help of a pair-wise comment ranking task. Our experiment reveals the challenges of this task and we outline promising paths for future work. The data and our code are available for research purposes from: hpi.de/naumann/projects/repeatability/text-mining.html.
10.
Bornemann, L., Bleifuß, T., Kalashnikov, D.V., Naumann, F., Srivastava, D.: Natural Key Discovery in Wikipedia Tables. Proceedings of The World Wide Web Conference (WWW). pp. 2789–2795 (2020).
Wikipedia is the largest encyclopedia to date. Scattered among its articles, there is an enormous number of tables that contain structured, relational information. In contrast to database tables, these webtables lack metadata, making it difficult to automatically interpret the knowledge they harbor. The natural key is a particularly important piece of metadata, which acts as a primary key and consists of attributes inherent to an entity. Determining natural keys is crucial for many tasks, such as information integration, table augmentation, or tracking changes to entities over time. To address this challenge, we formally define the notion of natural keys and propose a supervised learning approach to automatically detect natural keys in Wikipedia tables using carefully engineered features. Our solution includes novel features that extract information from time (a table’s version history) and space (other similar tables). On a curated dataset of 1,000 Wikipedia table histories, our model achieves 80% F-measure, which is at least 20% more than all related approaches. We use our model to discover natural keys in the entire corpus of Wikipedia tables and provide the dataset to the community to facilitate future research.
11.
Bejnordi, A.E., Krestel, R.: Dynamic Channel and Layer Gating in Convolutional Neural Networks. Proceedings of the 43rd German Conference on Artificial Intelligence (KI 2020) (2020).
Convolutional neural networks (CNN) are getting more and more complex, needing enormous computing resources and energy. In this paper, we propose methods for conditional computation in the context of image classification that allows a CNN to dynamically use its channels and layers conditioned on the input. To this end, we combine light-weight gating modules that can make binary decisions without causing much computational overhead. We argue, that combining the recently proposed channel gating mechanism with layer gating can significantly reduce the computational cost of large CNNs. Using discrete optimization algorithms, the gating modules are made aware of the context in which they are used and decide whether a particular channel and/or a particular layer will be executed. This results in neural networks that adapt their own topology conditioned on the input image. Experiments using the CIFAR10 and MNIST datasets show how competitive results in image classification with respect to accuracy can be achieved while saving up to 50% computational resources.
12.
Ehmüller, J., Kohlmeyer, L., McKee, H., Paeschke, D., Repke, T., Krestel, R., Naumann, F.: Sense Tree: Discovery of New Word Senses with Graph-based Scoring. Proceedings of the Conference on "Lernen, Wissen, Daten, Analysen" (LWDA). pp. 1–12 (2020).
Language is dynamic and constantly evolving: both the us-age context and the meaning of words change over time. Identifying words that acquired new meanings and the point in time at which new word senses emerged is elementary for word sense disambiguation and entity linking in historical texts. For example, cloud once stood mostly for the weather phenomenon and only recently gained the new sense of cloud computing. We propose a clustering-based approach that computes sense trees, showing how meanings of words change over time. The produced results are easy to interpret and explain using a drill-down mechanism. We evaluate our approach qualitatively on the Corpus of Historic American English (COHA), which spans two hundred years.
13.
Risch, J., Krestel, R.: Top Comment or Flop Comment? Predicting and Explaining User Engagement in Online News Discussions. Proceedings of the International Conference on Web and Social Media (ICWSM). pp. 579–589. AAAI (2020).
Comment sections below online news articles enjoy growing popularity among readers. However, the overwhelming number of comments makes it infeasible for the average news consumer to read all of them and hinders engaging discussions. Most platforms display comments in chronological order, which neglects that some of them are more relevant to users and are better conversation starters. In this paper, we systematically analyze user engagement in the form of the upvotes and replies that a comment receives. Based on comment texts, we train a model to distinguish comments that have either a high or low chance of receiving many upvotes and replies. Our evaluation on user comments from TheGuardian.com compares recurrent and convolutional neural network models, and a traditional feature-based classifier. Further, we investigate what makes some comments more engaging than others. To this end, we identify engagement triggers and arrange them in a taxonomy. Explanation methods for neural networks reveal which input words have the strongest influence on our model's predictions. In addition, we evaluate on a dataset of product reviews, which exhibit similar properties as user comments, such as featuring upvotes for helpfulness.
14.
Jain, N., Bartz, C., Krestel, R.: Automatic Matching of Paintings and Descriptions in Art-Historic Archives using Multimodal Analysis. 1st International Workshop on Artificial Intelligence for Historical Image Enrichment and Access (AI4HI-2020), co-located with LREC 2020 conference (2020).
Cultural heritage data plays a pivotal role in the understanding of human history and culture. A wealth of information is buried in art-historic archives which can be extracted via their digitization and analysis. This information can facilitate search and browsing, help art historians to track the provenance of artworks and enable wider semantic text exploration for digital cultural resources. However, this information is contained in images of artworks as well as textual descriptions, or annotations accompanied with the images. During the digitization of such resources, the valuable associations between the images and texts are frequently lost. In this project description, we propose an approach to retrieve the associations between images and texts for artworks from art-historic archives. To this end, we use machine learning to generate text descriptions for the extracted images on the one hand, and to detect descriptive phrases and titles of images from the text on the other hand. Finally, we use embeddings to align both, the descriptions and the images.
15.
Jain, N., Krestel, R.: Learning Fine-Grained Semantics for Multi-Relational Data. International Semantic Web Conference, 2020 Posters and Demos (2020).
The semantics of relations play a central role in the understanding and analysis of multi-relational data. Real-world relational datasets represented by knowledge graphs often contain polysemous relations between different types of entities, that represent multiple semantics. In this work, we present a data-driven method that can automatically discover the distinct semantics associated with high-level relations and derive an optimal number of sub-relations having fine-grained meaning. To this end, we perform clustering over vector representations of entities and relations obtained from knowledge graph embedding models.
16.
Risch, J., Krestel, R.: Toxic Comment Detection in Online Discussions. In: Agarwal, B., Nayak, R., Mittal, N., and Patnaik, S. (eds.) Deep Learning-Based Approaches for Sentiment Analysis. pp. 85–109. Springer (2020).
With the exponential growth in the use of social media networks such as Twitter, Facebook, and many others, an astronomical amount of big data has been generated. A substantial amount of this user-generated data is in form of text such as reviews, tweets, and blogs that provide numerous challenges as well as opportunities to NLP (Natural Language Processing) researchers for discovering meaningful information used in various applications. Sentiment analysis is the study that analyses people’s opinion and sentiment towards entities such as products, services, person, organisations etc. present in the text. Sentiment analysis and opinion mining is the most popular and interesting research problem. In recent years, Deep Learning approaches have emerged as powerful computational models and have shown significant success to deal with a massive amount of data in unsupervised settings. Deep learning is revolutionizing because it offers an effective way of learning representation and allows the system to learn features automatically from data without the need of explicitly designing them. Deep learning algorithms such as deep autoencoders, convolutional and recurrent neural networks (CNN) (RNN), Long Short-Term Memory (LSTM) and Generative Adversarial Networks (GAN) have reported providing significantly improved results in various natural language processing tasks including sentiment analysis.
17.
Risch, J., Ruff, R., Krestel, R.: Explaining Offensive Language Detection. Journal for Language Technology and Computational Linguistics (JLCL). 34, 29–47 (2020).
publisher: German Society for Computational Linguistics and Language Technology (GSCL)
Machine learning approaches have proven to be on or even above human-level accuracy for the task of offensive language detection. In contrast to human experts, however, they often lack the capability of giving explanations for their decisions. This article compares four different approaches to make offensive language detection explainable: an interpretable machine learning model (naive Bayes), a model-agnostic explainability method (LIME), a model-based explainability method (LRP), and a self-explanatory model (LSTM with an attention mechanism). Three different classification methods: SVM, naive Bayes, and LSTM are paired with appropriate explanation methods. To this end, we investigate the trade-off between classification performance and explainability of the respective classifiers. We conclude that, with the appropriate explanation methods, the superior classification performance of more complex models is worth the initial lack of explainability.
18.
Schirmer, P., Papenbrock, T., Koumarelas, I., Naumann, F.: Efficient Discovery of Matching Dependencies. ACM Transactions on Database Systems (TODS). 45, 1–33 (2020).
19.
Hameed, M., Naumann, F.: Data Preparation: A Survey of Commercial Tools. SIGMOD Record. 49, (2020).
Raw data are often messy: they follow different encodings, records are not well structured, values do not adhere to patterns, etc. Such data are in general not fit to be ingested by downstream applications, such as data analytics tools, or even by data management systems. The act of obtaining information from raw data relies on some data preparation process. Data preparation is integral to advanced data analysis and data management, not only for data science but for any data-driven applications. Existing data preparation tools are operational and useful, but there is still room for improvement and optimization. With increasing data volume and its messy nature, the demand for prepared data increases day by day. To cater to this demand, companies and researchers are developing techniques and tools for data preparation. To better understand the available data preparation systems, we have conducted a survey to investigate (1) prominent data preparation tools, (2) distinctive tool features, (3) the need for preliminary data processing even for these tools and, (4) features and abilities that are still lacking. We conclude with an argument in support of automatic and intelligent data preparation beyond traditional and simplistic techniques.
20.
Kruse, S., Kaoudi, Z., Quiane-Ruiz, J.-A., Chawla, S., Naumann, F., Contreras-Rojas, B.: RHEEMix in the Data Jungle: A Cost-based Optimizer for Cross-Platform Systems. VLDB Journal. 29, 1287–1310 (2020).
21.
Risch, J., Alder, N., Hewel, C., Krestel, R.: PatentMatch: A Dataset for Matching Patent Claims with Prior Art. ArXiv e-prints 2012.13919. (2020).
Patent examiners need to solve a complex information retrieval task when they assess the novelty and inventive step of claims made in a patent application. Given a claim, they search for prior art, which comprises all relevant publicly available information. This time-consuming task requires a deep understanding of the respective technical domain and the patent-domain-specific language. For these reasons, we address the computer-assisted search for prior art by creating a training dataset for supervised machine learning called PatentMatch. It contains pairs of claims from patent applications and semantically corresponding text passages of different degrees from cited patent documents. Each pair has been labeled by technically-skilled patent examiners from the European Patent Office. Accordingly, the label indicates the degree of semantic correspondence (matching), i.e., whether the text passage is prejudicial to the novelty of the claimed invention or not. Preliminary experiments using a baseline system show that PatentMatch can indeed be used for training a binary text pair classifier on this challenging information retrieval task. The dataset is available online: https://hpi.de/naumann/s/patentmatch
22.
Birnick, J., Bläsius, T., Friedrich, T., Naumann, F., Papenbrock, T., Schirneck, M.: Hitting Set Enumeration with Partial Information for Unique Column Combination Discovery. Proceedings of the VLDB Endowment. 13, 2270–2283 (2020).
Unique column combinations (UCCs) are a fundamental concept in relational databases. They identify entities in the data and support various data management activities. Still, UCCs are usually not explicitly defined and need to be discovered. State-of-the-art data profiling algorithms are able to efficiently discover UCCs in moderately sized datasets, but they tend to fail on large and, in particular, on wide datasets due to run time and memory limitations. In this paper, we introduce HPIValid, a novel UCC discovery algorithm that implements a faster and more resource-saving search strategy. HPIValid models the metadata discovery as a hitting set enumeration problem in hypergraphs. In this way, it combines efficient discovery techniques from data profiling research with the most recent theoretical insights into enumeration algorithms. Our evaluation shows that HPIValid is not only orders of magnitude faster than related work, it also has a much smaller memory footprint.
23.
Jiang, L., Naumann, F.: Holistic Primary Key and Foreign Key Detection. Journal of Intelligent Information Systems. 54, 439–461 (2020).
Primary keys (PKs) and foreign keys (FKs) are important elements of relational schemata in various applications, such as query optimization and data integration. However, in many cases, these constraints are unknown or not documented. Detecting them manually is time-consuming and even infeasible in large-scale datasets. We study the problem of discovering primary keys and foreign keys automatically and propose an algorithm to detect both, namely Holistic Primary Key and Foreign Key Detection (HoPF). PKs and FKs are subsets of the sets of unique column combinations (UCCs) and inclusion dependencies (INDs), respectively, for which efficient discovery algorithms are known. Using score functions, our approach is able to effectively extract the true PKs and FKs from the vast sets of valid UCCs and INDs. Several pruning rules are employed to speed up the procedure. We evaluate precision and recall on three benchmarks and two real-world datasets. The results show that our method is able to retrieve on average 88% of all primary keys, and 91% of all foreign keys. We compare the performance of algor with two baseline approaches that both assume the existence of primary keys.
24.
Koumarelas, I., Papenbrock, T., Naumann, F.: MDedup: Duplicate Detection with Matching Dependencies. Proceedings of the VLDB Endowment (PVLDB). 13, 712–725 (2020).
Duplicate detection is an integral part of data cleaning and serves to identify multiple representations of same real-world entities in (relational) datasets. Existing duplicate detection approaches are effective, but they are also hard to parameterize or require a lot of pre-labeled training data. Both parameterization and pre-labeling are at least domain-specific if not dataset-specific, which is a problem if a new dataset needs to be cleaned. For this reason, we propose a novel, rule-based and fully automatic duplicate detection approach that is based on matching dependencies (MDs). Our system uses automatically discovered MDs, various dataset features, and known gold standards to train a model that selects MDs as duplicate detection rules. Once trained, the model can select useful MDs for duplicate detection on any new dataset. To increase the generally low recall of MD-based data cleaning approaches, we propose an additional boosting step. Our experiments show that this approach reaches up to 94% F-measure and 100% precision on our evaluation datasets, which are good numbers considering that the system does not require domain or target data-specific configuration.
25.
Koumarelas, I., Jiang, L., Naumann, F.: Data Preparation for Duplicate Detection. Journal of Data and Information Quality (JDIQ). 12, 1–24 (2020).
Data errors represent a major issue in most application workflows. Before any important task can take place, a certain data quality has to be guaranteed, by eliminating a number of different errors that may appear in data. Typically, most of these errors are fixed with data preparation methods, such as whitespace removal. However, the particular error of duplicate records, where multiple records refer to the same entity, is usually eliminated independently with specialized techniques. Our work is the first to bring these two areas together by applying data preparation operations under a systematic approach, prior to performing duplicate detection. Our process workflow can be summarized as follows: It begins with the user providing as input a sample of the gold standard, the actual dataset, and optionally some constraints to domain-specific data preparations, such as address normalization. The preparation selection operates in two consecutive phases. First, to vastly reduce the search space of ineffective data preparations, decisions are made based on the improvement or worsening of pair similarities. Second, using the remaining data preparations an iterative leave-one-out classification process removes preparations one by one and determines the redundant preparations based on the achieved area under the precision-recall curve (AUC-PR). Using this workflow, we manage to improve the results of duplicate detection up to 19% in AUC-PR.
26.
Repke, T., Krestel, R.: Exploration Interface for Jointly Visualised Text and Graph Data. International Conference on Intelligent User Interfaces Companion (IUI ’20). (2020).
Many large text collections exhibit graph structures, either inherent to the content itself or encoded in the metadata of the individual documents. Example graphs extracted from document collections are co-author networks, citation networks, or named-entity-co-occurrence networks. Furthermore, social networks can be extracted from email corpora, tweets, or social media. When it comes to visualising these large corpora, traditionally either the textual content or the network graph are used. We propose to incorporate both, text and graph, to not only visualise the semantic information encoded in the documents’ content but also the relationships expressed by the inherent network structure in a two-dimensional landscape. We illustrate the effectiveness of our approach with an exploration interface for different real world datasets.
27.
Hacker, P., Krestel, R., Grundmann, S., Naumann, F.: Explainable AI under Contract and Tort Law: Legal Incentives and Technical Challenges. Artificial Intelligence and Law. (2020).
28.
Repke, T., Krestel, R.: Visualising Large Document Collections by Jointly Modeling Text and Network Structure. Proceedings of the Joint Conference on Digital Libraries (JCDL). (2020).
Many large text collections exhibit graph structures, either inherent to the content itself or encoded in the metadata of the individual documents. Example graphs extracted from document collections are co-author networks, citation networks, or named-entity-cooccurrence networks. Furthermore, social networks can be extracted from email corpora, tweets, or social media. When it comes to visualising these large corpora, either the textual content or the network graph are used. In this paper, we propose to incorporate both, text and graph, to not only visualise the semantic information encoded in the documents' content but also the relationships expressed by the inherent network structure. To this end, we introduce a novel algorithm based on multi-objective optimisation to jointly position embedded documents and graph nodes in a two-dimensional landscape. We illustrate the effectiveness of our approach with real-world datasets and show that we can capture the semantics of large document collections better than other visualisations based on either the content or the network information.
29.
Caruccio, L., Deufemia, V., Naumann, F., Polese, G.: Discovering Relaxed Functional Dependencies based on Multi-attribute Dominance. Transactions on Knowledge and Data Engineering (TKDE). (2020).
With the advent of big data and data lakes, data are often integrated from multiple sources. Such integrated data are often of poor quality, due to inconsistencies, errors, and so forth. One way to check the quality of data is to infer functional dependencies (FDs). However, in many modern applications it might be necessary to extract properties and relationships that are not captured through FDs, due to the necessity to admit exceptions, or to consider similarity rather than equality of data values. Relaxed FDs (RFDs) have been introduced to meet these needs, but their discovery from data adds further complexity to an already complex problem, also due to the necessity of specifying similarity and validity thresholds. We propose DOMINO, a new discovery algorithm for RFDs that exploits the concept of dominance in order to derive similarity thresholds of attribute values while inferring RFDs. An experimental evaluation on real datasets demonstrates the discovery performance and the effectiveness of the proposed algorithm.
