Contributions published at Dynamic and Distributed Information Systems (Abraham Bernstein)

Knowledge graphs (KGs) are at the core of numerous applications and their importance is increasing. Yet, knowledge evolves and so do KGs. PubMed, a search engine that primarily provides access to medical publications, adds an estimated 500'000 new records per year - each having the potential to require updates to a medical KG, like the National Cancer Institute Thesaurus. Depending on the applications that use such a medical KG, some of these updates have possibly wide-ranging impact, while others have only local effects. Estimating the impact of a change ex-ante is highly important, as it might make KG-engineers aware of the consequences of their actions during editing or may be used to highlight the importance of a new fragment of knowledge to be added to the KG for some application. This research description proposes a unified methodology for predicting the impact of changes in evolving KGs and introduces an evaluation framework to assess the quality of these predictions.

The updates on knowledge graphs (KGs) affect the services built on top of them. However, changes are not all the same: some updates drastically change the result of operations based on knowledge graph content; others do not lead to any variation. Estimating the impact of a change ex-ante is highly important, as it might make KG engineers aware of the consequences of their action during KG editing or may be used to highlight the importance of a new fragment of knowledge to be added to the KG for some application. The main goal of this contribution is to offer a formalization of the problem. Additionally, it presents some preliminary experiments on three different datasets considering embeddings as operation.Results show that the estimation can reach AUCs of 0.85, suggesting the feasibility of this research.

Over the last few years, natural language interfaces (NLI) for databases have gained significant traction both in academia and industry. These systems use very different approaches as described in recent survey papers. However, these systems have not been systematically compared against a set of benchmark questions in order to rigorously evaluate their functionalities and expressive power. In this paper, we give an overview over 24 recently developed NLIs for databases. Each of the systems is evaluated using a curated list of ten sample questions to show their strengths and weaknesses. We categorize the NLIs into four groups based on the methodology they are using: keyword-, pattern-, parsing- and grammar-based NLI. Overall, we learned that keyword-based systems are enough to answer simple questions. To solve more complex questions involving subqueries, the system needs to apply some sort of parsing to identify structural dependencies. Grammar-based systems are overall the most powerful ones, but are highly dependent on their manually designed rules. In addition to providing a systematic analysis of the major systems, we derive lessons learned that are vital for designing NLIs that can answer a wide range of user questions.

null

null

null

null

The topology of animal transport networks contributes substantially to how fast and to what extent a disease can transmit between animal holdings. Therefore, public authorities in many countries mandate livestock holdings to report all movements of animals. However, the reported data often does not contain information about the exact sequence of transports, making it impossible to assess the effect of truck sharing and truck contamination on disease transmission. The aim of this study was to analyze the topology of the Swiss pig transport network by means of social network analysis and to assess the implications for disease transmission between animal holdings. In particular, we studied how additional information about transport sequences changes the topology of the contact network. The study is based on the official animal movement database in Switzerland and a sample of transport data from one transport company. The results show that the Swiss pig transport network is highly fragmented, which mitigates the risk of a large-scale disease outbreak. By considering the time sequence of transports, we found that even in the worst case, only 0.34% of all farm-pairs were connected within one month. However, both network connectivity and individual connectedness of farms increased if truck sharing and especially truck contamination were considered. Therefore, the extent to which a disease may be transmitted between animal holdings may be underestimated if we only consider data from the official animal movement database. Our results highlight the need for a comprehensive analysis of contacts between farms that includes indirect contacts due to truck sharing and contamination. As the nature of animal transport networks is inherently temporal, we strongly suggest the use of temporal network measures in order to evaluate individual and overall risk of disease transmission through animal transportation.

Technological advances allow progressively more autonomous systems to become part of our society. Such systems can be especially useful when time pressure and uncertainty are part of a decision-making process, e.g. in a security context. However, by using such system, there is a risk that the output of the system does not match ethical expectation, e.g. because a suboptimal solution is selected or collateral damage occurs. This has two implications. Firstly, the actual advice or action the system performs should be as we prefer it to be. Secondly, the user needs to perceive the system as an ethical and trustworthy partner in the decision-making process, to ensure the system is actually used. This project focuses on the latter, and contributes to the further elaboration of empirical issues raised by the White Paper “Evaluation Schema for the Ethical Use of Autonomous Robotic Systems in Security Applications”. While there has been research on autonomous systems and ethics, the field is still very much developing. To our knowledge, the following specific factors in this research have not been combined before: different levels of autonomy in search and rescue scenarios, uncertainty and time pressure in ethical decision-making, and trust. In order to investigate the interplay of those factors, we use a multidisciplinary and experimental approach. Compared to standard experimental ethics that is usually vignette-based, we will present morally challenging scenarios to participants in a simulation. This setting allows more immersion into the ethical scenario and adds the human interaction component, which is important to research the perception and expectations of the user. Currently, an experimental setup is designed together with a simulation prototype; the experiment is going to take place with search and rescue recruits of the Swiss army. They will participate in simulations involving the use of drones controlled by the participants in two setting: a rescue mission where a limited number can be saved and a prevention mission (bringing down a terror drone) where there will be some casualties. The system will either provide decision support for a given scenario or autonomously take a decision on what to do; the user only has a veto option. After each scenario, question will be asked on ethical acceptability, ethical responsibility and trust. At the conference, we will present results of pretesting of different scenarios and we will further outline our research program. The results of this research should ultimately shape guidelines on how to build ethically trustworthy autonomous systems.

null

null

null

Knowledge graph embedding aims to learn distributed representations for entities and relations, and are proven to be effective in many applications. Crossover interactions --- bi-directional effects between entities and relations --- help select related information when predicting a new triple, but hasn't been formally discussed before. In this paper, we propose CrossE, a novel knowledge graph embedding which explicitly simulates crossover interactions. It not only learns one general embedding for each entity and relation as in most previous methods, but also generates multiple triple specific embeddings for both of them, named interaction embeddings. We evaluate the embeddings on typical link prediction task and find that CrossE achieves state-of-the-art results on complex and more challenging datasets. Furthermore, we evaluate the embeddings from a new perspective --- giving explanations for predicted triples, which is important for real applications. In this work, explanations for a triple are regarded as reliable closed-paths between head and tail entity. Compared to other baselines, we show experimentally that CrossE is more capable of generating reliable explanations to support its predictions, benefiting from interaction embeddings.

null

null

With the upcoming of computing systems that process large amounts of data, securing sensitive data has become a lasting concern. Research has shown that with the help of other data sets, anoynimzed data sets can be reconstructed. Differential privacy is a construct to secure sensitive data. The data is perturbed with noise to ensure privacy. Differential privacy has shown to be very effective against attempts to reconstruct anonymized data sets in combination with other data sets. Originally used on static data sets, differential privacy was extended to data streams. Four papers that introduce various algorithms are examined in this thesis. The algorithms are often of theoretical nature. To evaluate the real effects of the algorithms on the results of data streams, they are implemented and compared to each other. The focus is on the performance in terms of accuracy that the algorithms deliver.

In major political events and discussions, recommender algorithms play a large role in shaping opinions and setting the agenda for more traditional news media. These algorithms are used pervasively in social networks, media platforms and search engines. They determine what information is shown to the user and in which order. Existing recommender systems focus on improving accuracy based on historic interaction-data, which has received criticism for being detrimental to the goals of improving user experience and information diversity. Their accuracy is measured in terms of predicting future user behavior based on past observation. The most popular algorithms suggest items to users based on the choices of similar users. However, these systems are observed to promote a narrow set of already-popular items that are similar to past choices of the user, or are liked by many users. This limits users' exposure to diverse viewpoints and potentially increases polarization. One reason for the lack of diversity in existing approaches is that they optimize for average accuracy and click rates. There are other aspects of user-experience like new information and diering viewpoints, that are not taken into account. Similarly, a high average accuracy does not necessarily mean that the algorithm is similarly accurate for dierent groups of users and niche items. Besides this, training data may be biased because of previous recommendations, or the data collection process. Although information diversity is important, its denition depends heavily on the specic application and it is hard to nd a general denition in the context of recommender systems. Nevertheless, there are denitions that pertain to some specic aspects of diversity, such as entropy, coverage, personalization, and surprisal. There is a need to design recommender algorithms that balance accuracy and diversity in order to deal with the aforementioned problems. To approach this problem, we rst investigated state-of-the-art algorithms and found that they promote already popular products. We also found that recommender diversity depends on the optimization parameters of the algorithms, a factor that had not received enough attention before. In this context, we developed a graph- based method that promotes niche items without sacricing accuracy. We also developed a probabilistic latent-factor model that signicantly improves the coverage and long-tail diversity of recommendations. Currently, we are using these insights to develop new machine learning algorithms for diverse news recommendation. We plan to and deploy and test them in a large-scale experiment involving multiple news producers and consumers.

Analyzing data streams is a vital task in data science. Often, data comes in different shapes such as triples, tuples, relations, or matrices. Traditional stream processing systems, however, only process data in one of these formats. To enable the processing of streams combining different shapes of data, we developed a system that parses SPARQL queries using the Apache Jena parser and transforms them to Apache Flink topologies. With a custom data type and tailored functions, we enabled the integration of matrices in Jena and therefore allowed to mix graphs, relational, and linear algebra in an RDF graph. This provided a proof of concept that queries may be written for static data and – with the usage of the streaming engine Flink – can easily be run on data streams, even if they contain multiple of the aforementioned types.