@article{65182, abstract = {{The aggregation of rating metrics in reputation systems is crucial for mitigating information overload by condensing customer rating distributions into singular valence scores. While platforms typically employ technical aggregation functions, such as the arithmetic mean to capture product quality, it remains unclear whether these functions align with customers' innate aggregation patterns. To address this knowledge gap, we designed a controlled economic decision experiment to elicit customers' aggregation principles by analyzing their product ranking decisions and contrasting these with various reference functions. Our findings indicate that, on average, customers aggregate rating information in accordance with the arithmetic mean. However, a granular analysis at the individual level reveals significant heterogeneity in aggregation behavior, with a substantial cluster exhibiting binary patterns that focus equally on negative (1-2 star) and positive (4-5 star) ratings. Additional clusters concentrate on negative feedback, particularly 1-star ratings or 1-2 star ratings collectively. Notably, these inherent aggregation patterns exhibit stability across variations in numerical information presentation and are not significantly influenced by individual characteristics, such as online shopping experience, risk attitudes, or demographics. These findings suggest that while the arithmetic mean captures average consumer behavior, platforms could benefit from offering customizable aggregation options to better cater to diverse user preferences for processing rating distributions. By doing so, platforms can enhance the effectiveness of their reputation systems and improve the overall quality of decision-making for consumers.}}, author = {{van Straaten, Dirk and Mir Djawadi, Behnud and Melnikov, Vitalik and Hüllermeier, Eyke and Fahr, René}}, journal = {{SSRN Electronic Journal}}, publisher = {{Elsevier BV}}, title = {{{Aggregation Processes in Customer Rating Systems - Insights from an Economic Decision Experiment}}}, doi = {{http://dx.doi.org/10.2139/ssrn.6201258}}, year = {{2026}}, } @inproceedings{32311, abstract = {{Testing is one of the most frequent means of quality assurance for software. Property-based testing aims at generating test suites for checking code against user-defined properties. Test input generation is, however, most often independent of the property to be checked, and is instead based on random or user-defined data generation.In this paper, we present property-driven unit testing of functions with numerical inputs and outputs. Alike property-based testing, it allows users to define the properties to be tested for. Contrary to property-based testing, it also uses the property for a targeted generation of test inputs. Our approach is a form of learning-based testing where we first of all learn a model of a given black-box function using standard machine learning algorithms, and in a second step use model and property for test input generation. This allows us to test both predefined functions as well as machine learned regression models. Our experimental evaluation shows that our property-driven approach is more effective than standard property-based testing techniques.}}, author = {{Sharma, Arnab and Melnikov, Vitaly and Hüllermeier, Eyke and Wehrheim, Heike}}, booktitle = {{Proceedings of the 10th IEEE/ACM International Conference on Formal Methods in Software Engineering (FormaliSE)}}, pages = {{113--123}}, publisher = {{IEEE}}, title = {{{Property-Driven Testing of Black-Box Functions}}}, year = {{2022}}, } @techreport{45616, abstract = {{Aggregation metrics in reputation systems are important for overcoming information overload. When using these metrics, technical aggregation functions such as the arithmetic mean are implemented to measure the valence of product ratings. However, it is unclear whether the implemented aggregation functions match the inherent aggregation patterns of customers. In our experiment, we elicit customers' aggregation heuristics and contrast these with reference functions. Our findings indicate that, overall, the arithmetic mean performs best in comparison with other aggregation functions. However, our analysis on an individual level reveals heterogeneous aggregation patterns. Major clusters exhibit a binary bias (i.e., an over-weighting of moderate ratings and under-weighting of extreme ratings) in combination with the arithmetic mean. Minor clusters focus on 1-star ratings or negative (i.e., 1-star and 2-star) ratings. Thereby, inherent aggregation patterns are neither affected by variation of provided information nor by individual characteristics such as experience, risk attitudes, or demographics.}}, author = {{van Straaten, Dirk and Melnikov, Vitalik and Hüllermeier, Eyke and Mir Djawadi, Behnud and Fahr, René}}, title = {{{Accounting for Heuristics in Reputation Systems: An Interdisciplinary Approach on Aggregation Processes}}}, volume = {{72}}, year = {{2021}}, } @inproceedings{19953, abstract = {{Current GNN architectures use a vertex neighborhood aggregation scheme, which limits their discriminative power to that of the 1-dimensional Weisfeiler-Lehman (WL) graph isomorphism test. Here, we propose a novel graph convolution operator that is based on the 2-dimensional WL test. We formally show that the resulting 2-WL-GNN architecture is more discriminative than existing GNN approaches. This theoretical result is complemented by experimental studies using synthetic and real data. On multiple common graph classification benchmarks, we demonstrate that the proposed model is competitive with state-of-the-art graph kernels and GNNs.}}, author = {{Damke, Clemens and Melnikov, Vitaly and Hüllermeier, Eyke}}, booktitle = {{Proceedings of the 12th Asian Conference on Machine Learning (ACML 2020)}}, editor = {{Jialin Pan, Sinno and Sugiyama, Masashi}}, keywords = {{graph neural networks, Weisfeiler-Lehman test, cycle detection}}, location = {{Bangkok, Thailand}}, pages = {{49--64}}, publisher = {{PMLR}}, title = {{{A Novel Higher-order Weisfeiler-Lehman Graph Convolution}}}, volume = {{129}}, year = {{2020}}, } @inproceedings{15007, author = {{Melnikov, Vitaly and Hüllermeier, Eyke}}, booktitle = {{Proceedings ACML, Asian Conference on Machine Learning (Proceedings of Machine Learning Research, 101)}}, title = {{{Learning to Aggregate: Tackling the Aggregation/Disaggregation Problem for OWA}}}, doi = {{10.1016/j.jmva.2019.02.017}}, year = {{2019}}, } @inproceedings{184, abstract = {{In this paper, we propose a framework for a class of learning problems that we refer to as “learning to aggregate”. Roughly, learning-to-aggregate problems are supervised machine learning problems, in which instances are represented in the form of a composition of a (variable) number on constituents; such compositions are associated with an evaluation, score, or label, which is the target of the prediction task, and which can presumably be modeled in the form of a suitable aggregation of the properties of its constituents. Our learning-to-aggregate framework establishes a close connection between machine learning and a branch of mathematics devoted to the systematic study of aggregation functions. We specifically focus on a class of functions called uninorms, which combine conjunctive and disjunctive modes of aggregation. Experimental results for a corresponding model are presented for a review data set, for which the aggregation problem consists of combining different reviewer opinions about a paper into an overall decision of acceptance or rejection.}}, author = {{Melnikov, Vitaly and Hüllermeier, Eyke}}, booktitle = {{Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML/PKDD 2016)}}, pages = {{756--771}}, title = {{{Learning to Aggregate Using Uninorms}}}, doi = {{10.1007/978-3-319-46227-1_47}}, year = {{2016}}, } @inproceedings{10223, author = {{Melnikov, Vitaly and Hüllermeier, Eyke}}, booktitle = {{European Conference on Machine Learning and Knowledge Discovery in Databases, Part II, Riva del Garda, Italy}}, pages = {{756--771}}, title = {{{Learning to aggregate using uninorms, in Proceedings ECML/PKDD-2016}}}, year = {{2016}}, }