Biomedical subjects

Gondy Leroy

Publications and source records attributed to Gondy Leroy.

5 recordsLinked to original sources

PubMed2006

An end user evaluation of query formulation and results review tools in three medical meta-search engines.

PURPOSE: Retrieving sufficient relevant information online is difficult for many people because they use too few keywords to search and search engines do not provide many support tools. To further complicate the search, users often ignore support tools when available. Our goal is to evaluate in a realistic setting when users use support tools and how they perceive these tools. METHODS: We compared three medical search engines with support tools that require more or less effort from users to form a query and evaluate results. We carried out an end user study with 23 users who were asked to find information, i.e., subtopics and supporting abstracts, for a given theme. We used a balanced within-subjects design and report on the effectiveness, efficiency and usability of the support tools from the end user perspective. CONCLUSIONS: We found significant differences in efficiency but did not find significant differences in effectiveness between the three search engines. Dynamic user support tools requiring less effort led to higher efficiency. Fewer searches were needed and more documents were found per search when both query reformulation and result review tools dynamically adjust to the user query. The query reformulation tool that provided a long list of keywords, dynamically adjusted to the user query, was used most often and led to more subtopics. As hypothesized, the dynamic result review tools were used more often and led to more subtopics than static ones. These results were corroborated by the usability questionnaires, which showed that support tools that dynamically optimize output were preferred.

Consumer Behavior↗

PubMed2005

Effects of information and machine learning algorithms on word sense disambiguation with small datasets.

Current approaches to word sense disambiguation use (and often combine) various machine learning techniques. Most refer to characteristics of the ambiguity and its surrounding words and are based on thousands of examples. Unfortunately, developing large training sets is burdensome, and in response to this challenge, we investigate the use of symbolic knowledge for small datasets. A naïve Bayes classifier was trained for 15 words with 100 examples for each. Unified Medical Language System (UMLS) semantic types assigned to concepts found in the sentence and relationships between these semantic types form the knowledge base. The most frequent sense of a word served as the baseline. The effect of increasingly accurate symbolic knowledge was evaluated in nine experimental conditions. Performance was measured by accuracy based on 10-fold cross-validation. The best condition used only the semantic types of the words in the sentence. Accuracy was then on average 10% higher than the baseline; however, it varied from 8% deterioration to 29% improvement. To investigate this large variance, we performed several follow-up evaluations, testing additional algorithms (decision tree and neural network), and gold standards (per expert), but the results did not significantly differ. However, we noted a trend that the best disambiguation was found for words that were the least troublesome to the human evaluators. We conclude that neither algorithm nor individual human behavior cause these large differences, but that the structure of the UMLS Metathesaurus (used to represent senses of ambiguous words) contributes to inaccuracies in the gold standard, leading to varied performance of word sense disambiguation techniques.

Algorithms↗

PubMed2005

Communication software using pictures for use with Pocket PCs.

Autism spectrum disorder has become one of the most prevalent developmental disorders. A difficulty with communication is one of the main impairments. We are developing a digital library of images that will be used to help children with autism communicate without the need for reading or writing skills. Images will be displayed on Pocket PCs to convey messages. We are currently developing and evaluating the first prototype.

Audiovisual Aids↗

PubMed2004

Using symbolic knowledge in the UMLS to disambiguate words in small datasets with a naïve Bayes classifier.

Current approaches to word sense disambiguation use and combine various machine-learning techniques. Most refer to characteristics of the ambiguous word and surrounding words and are based on hundreds of examples. Unfortunately, developing large training sets is time-consuming. We investigate the use of symbolic knowledge to augment machine-learning techniques for small datasets. UMLS semantic types assigned to concepts found in the sentence and relationships between these semantic types form the knowledge base. A naïve Bayes classifier was trained for 15 words with 100 examples for each. The most frequent sense of a word served as the baseline. The effect of increasingly accurate symbolic knowledge was evaluated in eight experimental conditions. Performance was measured by accuracy based on 10-fold cross-validation. The best condition used only the semantic types of the words in the sentence. Accuracy was then on average 10% higher than the baseline; however, it varied from 8% deterioration to 29% improvement. In a follow-up evaluation, we noted a trend that the best disambiguation was found for words that were the least troublesome to the human evaluators.

Abstracting and Indexing↗

PubMed2003

A shallow parser based on closed-class words to capture relations in biomedical text.

Natural language processing for biomedical text currently focuses mostly on entity and relation extraction. These entities and relations are usually pre-specified entities, e.g., proteins, and pre-specified relations, e.g., inhibit relations. A shallow parser that captures the relations between noun phrases automatically from free text has been developed and evaluated. It uses heuristics and a noun phraser to capture entities of interest in the text. Cascaded finite state automata structure the relations between individual entities. The automata are based on closed-class English words and model generic relations not limited to specific words. The parser also recognizes coordinating conjunctions and captures negation in text, a feature usually ignored by others. Three cancer researchers evaluated 330 relations extracted from 26 abstracts of interest to them. There were 296 relations correctly extracted from the abstracts resulting in 90% precision of the relations and an average of 11 correct relations per abstract.

Abstracting and Indexing↗