Search

In this study we use machine learning to perform explainable business sector prediction from financial statements. Financial statements are a valuable source of information on the financial state and performance of firms. Recently, large-scale data on financial statements has become available in the form of open data sets. Previous work on such data mainly focused on predicting fraud and bankruptcy. In this paper we devise a model for business sector prediction, which has several valuable applications, including automated error and fraud detection. In addition, such a predictive model may help in completing similar datasets with missing sector information. The proposed method employs a supervised learning approach based on random forests that addresses business sector prediction as a classification task. Using a dataset from the Netherlands Chamber of Commerce, containing over 1.5 million financial statements from Dutch companies, we created an adequately-performing model for business sector prediction. By assessing which features are instrumental in the final classification model, we found that a small number of attributes is crucial for predicting the majority of business sectors. Interestingly, in some cases the presence or absence of a feature was more important than the value itself. The resulting insights may also prove useful in accounting, where the relation between financial statements and characteristics of the company is a frequently studied topic.

A lot of research in social network mining is concerned with theories and methodologies for community discovery, pattern detection and network evolution, as well as behavioural analysis and anomaly (misbehaviour) detection. The MAISoN workshop focuses on the use of social network data and methods for building predictive models that can be used to uncover hidden and unexpected aspects of user-generated content in order to extract actionable insights. The objective is to explore ways in which insights can be transformed into effective actions that can help organizations improve and refine their activities. Thus, the focus is on social network analysis and mining techniques for gaining actionable real-world insights. The 3rd International Workshop on Mining Actionable Insights from Social Networks (MAISoN 2019) was a half day workshop co-located with ICTIR 2019, the 5th ACM SIGIR International Conference on the Theory of Information Retrieval which took place from October 2 to 5, 2019 in Santa Clara, California, United States.

This paper proposes a novel approach to efficiently compute the exact closeness centrality values of all nodes in dynamically evolving directed and weighted networks. Closeness centrality is one of the most frequently used centrality measures in the field of social network analysis. It uses the total distance to all other nodes to determine node centrality. Previous work has addressed the problem of dynamically updating closeness centrality values for either undirected networks or only for the top- $k$ nodes in terms of closeness centrality. Here, we propose a fast approach for exactly computing all closeness centrality values at each timestamp of directed and weighted evolving networks. Such networks are prevalent in many real-world situations. The main ingredients of our approach are a combination of work filtering methods and efficient incremental updates that avoid unnecessary recomputation. We tested the approach on several real-world datasets of dynamic small-world networks and found that we have mean speed-ups of about 33 times. In addition, the method is highly parallelizable.

In this paper we present a novel approach to determine the exact diameter (longest shortest path length) of large graphs, in particular of the nowadays frequently studied small world networks. Typical examples include social networks, gene networks, web graphs and internet topology networks. Due to complexity issues, the diameter is often calculated based on a sample of only a fraction of the nodes in the graph, or some approximation algorithm is applied. We instead propose an exact algorithm that uses various lower and upper bounds as well as effective node selection and pruning strategies in order to evaluate only the critical nodes which ultimately determine the diameter. We will show that our algorithm is able to quickly determine the exact diameter of various large datasets of small world networks with millions of nodes and hundreds of millions of links, whereas before only approximations could be given.