Software Product OverviewAll Talking Dolphin software is built in 100% pure Java, and is available for commercial license. Classification Software ClassifyThis! is a new statistical classification software suite, enabling rapid development of automatic document classification systems with advanced statistical machine learning technology. It is available now for commercial license from Talking Dolphin. Clustering SoftwareTalking Dolphin's proprietary document clustering software is currently available as a beta release. Contact Talking Dolphin to learn more. Coming SoonTalking Dolphin has a number of other products in the pipeline, including:
©2008 Talking Dolphin, LLC |
ClassifiersClassifiers are software programs capable of making simple decisions automatically. For instance, they may be capable of deciding whether an email is spam or not, or what category a new product belongs to. Machine learning techniques allow us to learn classifiers automatically from annotated training data. Statistical classifiers are classifiers that represent quantitative statistics about the problem, and can combine "soft" evidence to make decisions. ClusteringClustering is the process of automatically partitioning unlabeled data elements into a finite set of groups. It is the most basic example of unsupervised machine learning, in which the goal is to automatically recognize patterns in unannotated data. Clustering can be useful as a first step in organizing large collections of heterogeneous documents: for example an email box containing many different types of mail. Information ExtractionInformation extraction is the process of automatically extracting structured information from unstructured or semi-structured text documents. One of the most common information extraction tasks is named entity recognition, or extracting the names of people, places, and organizations that are mentioned in an unstructured news article. |