1,721,006 research outputs found
Supervised and unsupervised learning techniques for profiling SAGE results.
No full textUsing Serial Analysis it is now possible to obtain quantita tive measurements of the expression of thousands of genes present in a biological sample. Serial analysis yield a global view of gene expression that can be used in a number of interesting ways.
In this paper we are investigating two different approaches for analyz ing the analysis of data obtained from SAGE experiments. The first one is a supervised learning process: a classification of cancer tissue using decision trees and Support Vector Machines (SVM). After that, we will analyze the results achieved by a unsupervised learning method: hierar chical clustering. Finally, we tried to characterize the groups found by clustering, using the classification techniques cited before
La sedicesima conferenza europea di intelligenza artificiale (ECAI)
No full textRelazione sulla sedicesima conferenza europea di intelligenza artificiale (ECAI
Exploiting Association and Correlation Rules Parameters for Learning Bayesian Networks
No full textIn data mining, association and correlation rules
are inferred from data in order to highlight statistical dependencies among attributes. The metrics defined for evaluating these rules can be exploited to score relationships between attributes in Bayesian network learning. In this paper, we propose two novel methods for learning Bayesian networks from data that are
based on the K2 learning algorithm and that improve it by exploiting parameters
normally defined for association and correlation rules.
In particular, we propose the algorithms K2-Lift and K2-X2, that exploit the lift metric and the X2 metric respectively. We compare
K2-Lift, K2-X2 with K2 on artificial data and on
three test Bayesian networks. The experiments show that both our algorithms improve K2 with respect to the quality of the
learned network. Moreover, a comparison of K2-Lift and K2-X2 with a genetic algorithm approach on two benchmark networks show superior results on one network and comparable results on the other
Application of machine learning techniques for the forecasting of fashion trends
No full textNell’articolo si mostra come sia possibile ed utile applicare
tecniche di Intelligenza Artificiale in un settore altamente
creativo come quello della moda, con particolare riferimento
alla previsione dei “trend”.
Il lavoro descrive un prototipo di sistema basato sulla
conoscenza in grado di generare automaticamente le previsioni
dei “trend” nel settore della moda. Nel nostro caso,
ogni “trend” e’ rappresentato dai colori utilizzabili per le
nuove collezioni che meglio rappresentano le parole chiave
e le immagini sulle quali lo stilista ha deciso di incentrare
una collezione.
Per la rappresentazione in modelli della conoscenza
utilizzata dal sistema per le previsioni sono state sperimentate
due diverse metodologie: le reti Bayesiane e gli
alberi decisionali. Questi modelli sono generati mediante
l’utilizzo di tecniche di apprendimento automatico a
partire da dataset di previsioni effettuate negli anni passati.
Nell’articolo vengono presentati il modo con cui utilizzare
tali modelli per la previsione e gli esperimenti svolti al
fine di valutare la loro performance
Finding biological process modifications in cancer tissues by mining gene expression correlations
No full textAbstract Background Through the use of DNA microarrays it is now possible to obtain quantitative measurements of the expression of thousands of genes from a biological sample. This technology yields a global view of gene expression that can be used in several ways. Functional insight into expression profiles is routinely obtained by using Gene Ontology terms associated to the cellular genes. In this paper, we deal with functional data mining from expression profiles, proposing a novel approach that studies the correlations between genes and their relations to Gene Ontology (GO). By using this "functional correlations comparison" we explore all possible pairs of genes identifying the affected biological processes by analyzing in a pair-wise manner gene expression patterns and linking correlated pairs with Gene Ontology terms. Results We apply here this "functional correlations comparison" approach to identify the existing correlations in hepatocarcinoma (161 microarray experiments) and to reveal functional differences between normal liver and cancer tissues. The number of well-correlated pairs in each GO term highlights several differences in genetic interactions between cancer and normal tissues. We performed a bootstrap analysis in order to compute false detection rates (FDR) and confidence limits. Conclusion Experimental results show the main advantage of the applied method: it both picks up general and specific GO terms (in particular it shows a fine resolution in the specific GO terms). The results obtained by this novel method are highly coherent with the ones proposed by other cancer biology studies. But additionally they highlight the most specific and interesting GO terms helping the biologist to focus his/her studies on the most relevant biological processes.</p
Discovering Validation Rules from Microbiological Data
No full textA huge amount of data is daily collected from clinical mi-
crobiology laboratories. These data concern the resistance or susceptibil-
ity of bacteria to tested antibiotics. Almost all microbiology laboratories
follow standard antibiotic testing guidelines which suggest antibiotic test
execution methods and result interpretation and validation (among them,
those annually published by NCCLS 2)3)). Guidelines basically specify, for
each species, the antibiotics to be tested, how to interpret the results of
tests and a list of exceptions regarding particular antibiotic test results.
Even if these standards are quite assessed, they do not consider pecu-
liar features of a given hospital laboratory, which possibly influence the
antimicrobial test results, and the further validation process.
In order to improve and better tailor the validation process, we have
applied knowledge discovery techniques, and data mining in particular,
to microbiological data with the purpose of discovering new validation
rules, not yet included in NCCLS guidelines, but considered plausible and
correct by interviewed experts. In particular, we applied the knowledge
discovery process in order to find (association) rules relating to each other
the susceptibility or resistance of a bacterium to different antibiotics.
This approach is not antithetic, but complementary to that based on
NCCLS rules: it proved very effective in validating some of them, and
also in extending that compendium. In this respect, the new discovered
knowledge has lead microbiologists to be aware of new correlations among
some antimicrobial test results, which were previously unnoticed. Last
but not least, the new discovered rules, taking into account the history
of the considered laboratory, are better tailored to the hospital situation,
and this is very important since some resistances to antibiotics are specific
to particular, local hospital environments
Rule-based Programming for Building Expert Systems: a Comparison in the Microbiological Data Validation and Surveillance Domain
No full textIn this work, we compare three rule-based programming tools used for building an
expert system for microbiological laboratory data validation and bacteria infections
monitoring. The first prototype of the system was implemented in KAPPA-PC. We
report on the implementation and performance by comparing KAPPA-PC with two
other more recent tools, namely JESS and ILOG JRULES. In order to test each
tool we realized three simple test applications capable to perform some tasks that
are peculiar of our expert system
Marker Analysis with APRIORI-Based Algorithms
No full textIn genetic studies, polygenic diseases are often analyzed searching for marker patterns that play a significant role in the susceptibility to the disease. In this paper we consider a dataset regarding periodontitis, that includes the analysis of nine genetic markers for 148 patients. We analyze these data by using two APRIORI-based algorithms: APRIORISD and APRIORI with filtering. The discovered rules (especially those found by APRIORI with filtering) confirmed the results published on
periodontitis
Combining apriori and bootstrap techniques for marker analysis
No full textIn genetic studies, complex diseases are often analyzed searching for marker patterns that play a significant role in the susceptibility to the disease.
In this paper we consider a dataset regarding periodontitis, that includes the analysis of nine genetic markers for 148 individuals.
We analyze these data by using a novel subgroup discovering algorithm, named APRIORI-B, that is based on APRIORI and bootstrap techniques. This algorithm can use different metrics for rule selection.
Experiments conducted by using as rule metrics novelty and confirmation, confirmed some previous results published on periodontitis
Learning specifications of interaction protocols and business processes and proving their properties
No full textIn this paper, we overview our recent research activity concerning
the induction of Logic Programming specifications, and the proof of their properties via Abductive Logic Programming. Both the inductive and abductive tools here briefly described have been applied to respectively learn and verify (properties of) interaction protocols in multi-agent systems, Web service choreographies, careflows and business processes
- …
