Application of Spreading Activation Techniques in InformationRetrieval
Artificial Intelligence Review
Probabilistic latent semantic indexing
Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval
ROCK: a robust clustering algorithm for categorical attributes
Information Systems
Self-Organizing Maps
COOLCAT: an entropy-based algorithm for categorical clustering
Proceedings of the eleventh international conference on Information and knowledge management
Similarity between words computed by spreading activation on an English dictionary
EACL '93 Proceedings of the sixth conference on European chapter of the Association for Computational Linguistics
Robust growing neural gas algorithm with application in cluster analysis
Neural Networks - 2004 Special issue: New developments in self-organizing systems
Compression and Machine Learning: A New Perspective on Feature Space Vectors
DCC '06 Proceedings of the Data Compression Conference
Data Mining: A Knowledge Discovery Approach
Data Mining: A Knowledge Discovery Approach
PCA and SVD with nonnegative loadings
Pattern Recognition
ePart '09 Proceedings of the 1st International Conference on Electronic Participation
Pure spreading activation is pointless
Proceedings of the 18th ACM conference on Information and knowledge management
Event Correlation on the Basis of Activation Patterns
PDP '10 Proceedings of the 2010 18th Euromicro Conference on Parallel, Distributed and Network-based Processing
From NLP (natural language processing) to MLP (machine language processing)
MMM-ACNS'10 Proceedings of the 5th international conference on Mathematical methods, models and architectures for computer network security
Node Similarities from Spreading Activation
ICDM '10 Proceedings of the 2010 IEEE International Conference on Data Mining
Data Mining: Practical Machine Learning Tools and Techniques
Data Mining: Practical Machine Learning Tools and Techniques
Extracting semantic knowledge from twitter
ePart'11 Proceedings of the Third IFIP WG 8.5 international conference on Electronic participation
Kernel k-means for categorical data
IDA'05 Proceedings of the 6th international conference on Advances in Intelligent Data Analysis
| Hi-index | 0.00 |
Machine learning algorithms play in an important role in knowledge discovery due to their capability to analyze data for which only limited a priori knowledge is available. Unfortunately, their application and the extraction of information from the trained algorithm models highly depend on the nature of the analyzed data and the algorithm models. Therefore, the deployment of knowledge discovery processes in heterogeneous domains causes the requirement for time-consuming process adaptations. In this work, we argue that the value-centric feature vector representation used within machine learning is the main reason for the necessity to create such highly domain-specific setups. Therefore, the Semantic Pattern Transformation is presented, which transforms the value-centric feature vectors into a semantic representation -- the Semantic Patterns. The principle idea behind this process is to analyze the relations (co-occurrences) of feature values within the value-centric feature vectors, and use this information within a new vector-based representation -- the Semantic Patterns. The simple model used within this representation comes with many advantages, including the simplified setup procedure for arbitrary machine learning algorithms within heterogeneous domains and the simple interpretation of the underlying model. We have already successfully applied the Semantic Pattern Transformation in numerous applications including text analysis, event correlation, malware detection, analysis of Twitter data and Android application analysis. This work extends the application-oriented works by a detailed explanation of the method and a thorough evaluation of supervised and unsupervised learning scenarios, and the application within semantic-aware search algorithms.