Publicaties Associatieve Netwerken

Applying Power Graph Analysis to Weighted Graphs (PDF)
We expanded Power Graph Analysis for use with weighted graphs, applying the technique to document categorisation with promising results. With the additional weight information we were able to create more accurate representations of the underlying data while maintaining a high level of edge reduction and improving visualisation of the graph.

Document Categorization using Multilingual Associative Networks based on Wikipedia (PDF)
Associative networks are a connectionist language model with the ability to categorize large sets of documents. In this research we combine monolingual associative networks based on Wikipedia to create a larger, multilingual associative network, using the cross-lingual connections between Wikipedia articles. We prove that such multilingual associative networks perform better than monolingual associative networks in tasks related to document categorization by comparing the results of both types of associative network on a multilingual dataset.

Hierarchical Document Categorization using Associative Networks (PDF)
Associative networks are a connectionist language model with the ability to handle dynamic data. We used two associative networks to categorize random sets of related Wikipedia articles with only their raw text as input. We then compared the resulting categorization to a gold standard: the manual categorization by Wikipedia authors and used a neural network as a baseline. We also determined a human rating by having a group of judges rank the four categorization methods by correctness and by usefulness with regards to finding information. Based on these tests, we determined that associative networks produce results that are clearly better than the neural network baseline, coming close to the gold standard in terms of usefulness and correctness. Furthermore, automated testing suggests these results continue to hold for large datasets.

Using Natural Language Processing to Improve Document Categorization with Associative Networks (PDF)
Associative networks are a connectionist language model with the ability to handle large sets of documents. In this research we investigated the use of natural language processing techniques (part-of-speech tagging and parsing) in combination with Associative Networks for document categorization and compare the results to a TF-IDF baseline. By filtering out unwanted observations and preselecting relevant data based on sentence structure, natural language processing can pre-filter information before it enters the associative network, thus improving results.

Using Wikipedia with Associative Networks for Document Classification (PDF)
We demonstrate a new technique for building associative networks based on Wikipedia, comparing them to WordNet-based associative networks that we used previously, finding the Wikipedia-based networks to perform better at document classification. Additionally, we compare the performance of associative networks to various other text classification techniques using the Reuters-21578 dataset, establishing that associative networks can achieve comparable results.