Proceedings of the ... International Conference on Machine Learning and Applications. International Conference on Machine Learning and Applications最新文献

In this study, we explore the feasibility of speech-based techniques to automatically evaluate a nonword repetition (NWR) test. NWR tests, a useful marker for detecting language impairment, require repetition of pronounceable nonwords, such as "D OY F", presented aurally by an examiner or via a recording. Our proposed method leverages ASR techniques to first transcribe verbal responses. Second, it applies machine learning techniques to ASR output for predicting gold standard scores provided by speech and language pathologists. Our experimental results for a sample of 101 children (42 with autism spectrum disorders, or ASD; 18 with specific language impairment, or SLI; and 41 typically developed, or TD) show that the proposed approach is successful in predicting scores on this test, with averaged product-moment correlations of 0.74 and mean absolute error of 0.06 (on a observed score range from 0.34 to 0.97) between observed and predicted ratings.

Bayesian network structures are usually built using only the data and starting from an empty network or from a naïve Bayes structure. Very often, in some domains, like medicine, a prior structure knowledge is already known. This structure can be automatically or manually refined in search for better performance models. In this work, we take Bayesian networks built by specialists and show that minor perturbations to this original network can yield better classifiers with a very small computational cost, while maintaining most of the intended meaning of the original model.

Given that unstructured data is increasing exponentially everyday, extracting and understanding the information, themes, and relationships from large collections of documents is increasingly important to researchers in many disciplines including biomedicine. Latent Dirichlet Allocation (LDA) is an unsupervised topic modeling technique based on the "bag-of-words" assumption that has been applied extensively to unveil hidden semantic themes within large sets of textual documents. Recently, it was extended using the "bag-of-n-grams" paradigm to account for word order. In this paper, we present an alternative phrase based LDA model to move from a bag of words or n-grams paradigm to a "bag-of-key-phrases" setting by applying a key phrase extraction technique, the C-value method, to further explore latent themes. We evaluate our approach by using a phrase intrusion user study and demonstrate that our model can help LDA generate better and more interpretable topics than those generated using the bag-of-n-grams approach. Given topic models essentially are statistical tools, an important problem in topic modeling is that of visualizing and interacting with the models to understand and extract new information from a collection. To evaluate our phrase based modeling approach in this context, we incorporate it in an open source interactive topic browser. Qualitative evaluations of this browser with biomedical experts demonstrate that our approach can aid biomedical researchers gain better and faster understanding of their document collections.

Large amount of electronic clinical data encompasses important information in free text format. To be able to help guide medical decision-making, text needs to be efficiently processed and coded. In this research, we investigate techniques to improve classification of Emergency Department computed tomography (CT) reports. The proposed system uses Natural Language Processing (NLP) to generate structured output from the reports and then machine learning techniques to code for the presence of clinically important injuries for traumatic orbital fracture victims. Topic modeling of the corpora is also utilized as an alternative representation of the patient reports. Our results show that both NLP and topic modeling improves raw text classification results. Within NLP features, filtering the codes using modifiers produces the best performance. Topic modeling shows mixed results. Topic vectors provide good dimensionality reduction and get comparable classification results as with NLP features. However, binary topic classification fails to improve upon raw text classification.

Given prior human judgments of the condition of an object it is possible to use these judgments to make a maximal likelihood estimate of what future human judgments of the condition of that object will be. However, if one has a reasonably large collection of similar objects and the prior human judgments of a number of judges regarding the condition of each object in the collection, then it is possible to make predictions of future human judgments for the whole collection that are superior to the simple maximal likelihood estimate for each object in isolation. This is possible because the multiple judgments over the collection allow an analysis to determine the relative value of a judge as compared with the other judges in the group and this value can be used to augment or diminish a particular judge's influence in predicting future judgments. Here we study and compare five different methods for making such improved predictions and show that each is superior to simple maximal likelihood estimates.