http://dl.gi.de/handle/20.500.12116/35769 Tue, 21 Jul 2026 13:29:35 GMT 2026-07-21T13:29:35Z http://dl.gi.de:80/bitstream/id/dd3b695f-e749-4c7d-93ea-e42ae5a9d3be/ http://dl.gi.de/handle/20.500.12116/35769 http://dl.gi.de/handle/20.500.12116/35782 Quantitative comparison of polarity lexicons in sentiment analysis tasks: Using a lexicon overlap score for similarity measurement between lexicons Welter, Felix J.M. Becker, Michael Sentiment classification is either based on sentiment lexicons or machine learning. For the construction and improvement of sentiment lexicons, several approaches and algorithms have been designed. The resulting lexicons are commonly benchmarked in different tasks and compared by their respective performance. However, this measure depends on the application domain. This work proposes a method for context-independent comparison of sentiment lexicons. Three scoring methods for similarity measurement of lexicons are explained. Furthermore, exemplarily applications of the scores are shown, including lexicon similarity analysis before and after expansion via a Distributional Thesaurus and clustering of lexicons. Adaptability and limitations of the lexicon overlap score and the demonstrated applications are discussed. Wed, 01 Jan 2020 00:00:00 GMT http://dl.gi.de/handle/20.500.12116/35782 2020-01-01T00:00:00Z http://dl.gi.de/handle/20.500.12116/35781 Semantic Code Search with Neural Bag-of-Words and Graph Convolutional Networks Sieper, Anna Abad; Amarkhel, Omar; Diez, Savina; Petrak, Dominic Becker, Michael Software developers are often confronted with tasks for which there are widespread solution patterns. Searching for solutions using natural language queries often leads to unsatisfying results. Github, Microsoft Research and Weights & Biases created the CodeSearchNet Challenge to address this problem. Its goal is to develop code search approaches that return the code that best matches a natural language query. In this paper, we investigate two different approaches in this context. First, a Neural Bag-of-Words encoder using TF-IDF weighting and second, a Graph Convolutional Network which includes the call hierarchy in a target method’s representation. In our experiments we were able to improve the Neural Bag-of-Words models, whose results were published in the CodeSearchNet Challenge. Our Neural Bag-of-Words encoder improves the MRR by 4.38% for Python and 4.98% for Java. The Graph Convolutional Network did not improve the results over of the Neural Bag-of-Words model. Wed, 01 Jan 2020 00:00:00 GMT http://dl.gi.de/handle/20.500.12116/35781 2020-01-01T00:00:00Z http://dl.gi.de/handle/20.500.12116/35775 Nutzensteigernde Faktoren zur Optimierung von Mobile Payment für Senioren als hilfsbedürftige Anwendergruppe Möllers, Frederike; Oberthür, Tim Becker, Michael Mobile Zahlungsprozesse, die mit Hilfe eines mobilen Endgerätes durchgeführt und als Mobile Payment bezeichnet werden, finden bereits in weiten Teilen des E-Commerce und stationären Einzelhandels Anwendung. Die angebotenen Lösungen entsprechen dabei allerdings noch nicht den spezifischen Anforderungen aller potenziellen Anwendergruppen. So haben Senioren, welche möglicherweise körperlich oder geistig eingeschränkt sind und damit eine hilfsbedürftige Anwendergruppe darstellen, besondere Anforderungen an Mobile Payment. In dieser Arbeit wird Mobile Payment aus Prozessperspektive betrachtet und es werden unter Berücksichtigung der besonderen Eigenschaften und Bedürfnissen von Senioren als hilfsbedürftige Anwendergruppe nutzensteigernde Faktoren definiert. Neben der Automatisierung bislang manuell aufwändiger Teilschritte mobiler Zahlungsverfahren lassen sich unter anderem die Verbreitung und Akzeptanz innerhalb der Anwendergruppe und im gesamten sozio-ökonomischen System als solche Faktoren identifizieren. Wed, 01 Jan 2020 00:00:00 GMT http://dl.gi.de/handle/20.500.12116/35775 2020-01-01T00:00:00Z http://dl.gi.de/handle/20.500.12116/35780 Developing a game AI for Murus Gallicus Wilson, Philip Wilson; Savinov, Andrej; Kadavanich, Annabella Becker, Michael The development of game AIs has been a popular challenge in the last years. One of the best game agents, AlphaZero, was developed by DeepMind in 2017 and superseded by MuZero in 2019. Both agents are based on algorithms that perfectly learn to play any game within not even a day, given they are fed the game’s rules. The development of such game AIs does not necessarily require big computation centers like the ones Google has. In this work, we show how to develop and implement a Murus Gallicus game AI using mainly GOFAI (Good Old-Fashioned Artificial Intelligence) methods. We start with a comparison between different search tree algorithms, including MiniMax, NegaMax, NegaScout (principal variation search) and show how transposition tables can be used for optimization. Furthermore, we demonstrate the advantages of a dynamic value function and time management while searching for the best move. Lastly, we evaluate the application of Evolutionary Learning (EL), explaining how we trained specific parameters. Wed, 01 Jan 2020 00:00:00 GMT http://dl.gi.de/handle/20.500.12116/35780 2020-01-01T00:00:00Z