Particle Competition and Cooperation in Networks for Semi-Supervised Learning with Concept Drift

Concept drift is a problem of increasing importance in machine learning and data mining. Data sets under analysis are no longer only static databases, but also data streams in which concepts and data distributions may not be stable over time. However, most learning algorithms produced so far are based on the assumption that data comes from a fixed distribution, so they are not suitable to handle concept drifts. Moreover, some concept drifts applications requires fast response, which means an algorithm must always be (re) trained with the latest available data. But the process of labeling data is usually expensive and/or time consuming when compared to unlabeled data acquisition, thus only a small fraction of the incoming data may be effectively labeled. Semi-supervised learning methods may help in this scenario, as they use both labeled and unlabeled data in the training process. However, most of them are also based on the assumption that the data is static. Therefore, semi-supervised learning with concept drifts is still an open challenge in machine learning. Recently, a particle competition and cooperation approach was used to realize graph-based semi-supervised learning from static data. In this paper, we extend that approach to handle data streams and concept drift. The result is a passive algorithm using a single classifier, which naturally adapts to concept changes, without any explicit drift detection mechanism. Its built-in mechanisms provide a natural way of learning from new data, gradually forgetting older knowledge as older labeled data items became less influent on the classification of newer data items. Some computer simulation are presented, showing the effectiveness of the proposed method.

The assessment of the quality of sugar using electronic tongue and machine learning algorithms

The correct classification of sugar according to its physico-chemical characteristics directly influences the value of the product and its acceptance by the market. This study shows that using an electronic tongue system along with established techniques of supervised learning leads to the correct classification of sugar samples according to their qualities. In this paper, we offer two new real, public and non-encoded sugar datasets whose attributes were automatically collected using an electronic tongue, with and without pH controlling. Moreover, we compare the performance achieved by several established machine learning methods. Our experiments were diligently designed to ensure statistically sound results and they indicate that k-nearest neighbors method outperforms other evaluated classifiers and, hence, it can be used as a good baseline for further comparison. © 2012 IEEE.

Influence of agrochemicals fipronil and imidacloprid on the learning behavior of Apis mellifera L. honeybees

Agrochemicals on crop cultivated areas is a source of contamination for bees and may cause physiological and behavioral disorders and mortality. The LD50 of the pesticides fipronil and imidacloprid was determined and their effect on the learning behavior of Apis mellifera L. honeybee evaluated. LD50 was determined by the ingestion of contaminated food with different concentrations of insecticide concentrations: Fipronil (0, 0.8, 0.4, 0.2, 0.1 and 0.05 µg bee-1) and imidacloprid (0, 0.4, 0.2, 0.1, 0.05 and 0.025 µg bee-1). The method of proboscis extension reflection (PER) and learning through citral odor evaluated their responses to food stimulation. LD50 obtained were 0.28 ± 0.11 and 0.10 ± 0.04 µg bee-1 for fipronil and imidacloprid, respectively. The PER test showed no significant difference (p < 0.05) although agrochemicals affected the learning of bees. Insecticides fipronil and imidacloprid are extremely harmful to foraging Africanized Apis mellifera bees.

Unsupervised manifold learning using Reciprocal kNN Graphs in image re-ranking and rank aggregation tasks

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

A method using artificial neural networks to morphologically assess mouse blastocyst quality

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Auditory evoked potential: a proposal for further evaluation in children with learning disabilities

The information presented in this paper demonstrates the author's experience in previews cross-sectional studies conducted in Brazil, in comparison with the current literature. Over the last 10 years, auditory evoked potential (AEP) has been used in children with learning disabilities. This method is critical to analyze the quality of the processing in time and indicates the specific neural demands and circuits of the sensorial and cognitive process in this clinical population. Some studies with children with dyslexia and learning disabilities were shown here to illustrate the use of AEP in this population.

Particle competition and cooperation for semi-supervised learning with label noise

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Learning HMMs for nucleotide sequences from amino acid alignments

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Supervised variational relevance learning, an analytic geometric feature selection with applications to omic datasets

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Image segmentation through combined methods: watershed transform, unsupervised distance learning and normalized cut

Research on image processing has shown that combining segmentation methods may lead to a solid approach to extract semantic information from different sort of images. Within this context, the Normalized Cut (NCut) is usually used as a final partitioning tool for graphs modeled in some chosen method. This work explores the Watershed Transform as a modeling tool, using different criteria of the hierarchical Watershed to convert an image into an adjacency graph. The Watershed is combined with an unsupervised distance learning step that redistributes the graph weights and redefines the Similarity matrix, before the final segmentation step using NCut. Adopting the Berkeley Segmentation Data Set and Benchmark as a background, our goal is to compare the results obtained for this method with previous work to validate its performance.

A self-learning simulated annealing algorithm for global optimizations of electromagnetic devices

A self-learning simulated annealing algorithm is developed by combining the characteristics of simulated annealing and domain elimination methods. The algorithm is validated by using a standard mathematical function and by optimizing the end region of a practical power transformer. The numerical results show that the CPU time required by the proposed method is about one third of that using conventional simulated annealing algorithm.

An Uphill Battle: Incorporating cooperative learning using a largely individualized curriculum

In this action research study of my classroom of 8th grade mathematics, I investigated if cooperative learning could be an effective teaching method with the Saxon curriculum. Saxon curriculum is largely individualized in that most lessons could be completed without much group interaction. I discovered that cooperative learning was very successful with the curriculum as long as it was structured. Ninety-five percent of the students in the study preferred to work in groups, and I observed mathematical communication grow with most of the students. As a result of this research, I plan to continue to incorporate cooperative learning into my mathematics classroom. I will use cooperative learning with all of my mathematics classes, even the ones that do not use the Saxon curriculum. I believe in the power of working together.

Cooperative Learning in Relation to Problem Solving in the Mathematics Classroom

In this action research study of my classroom of 5th grade mathematics, I investigated cooperative learning and how it is related to problem solving as well as written and oral communication. I discovered that cooperative learning has a positive impact on students’ abilities in problem solving and their overall impression of mathematics and group work. I also found that my students’ communication skills improved in oral explanations of their work. As a result of this research I plan to continue my implementation of cooperative learning in my classroom as a general method of teaching. I also plan to continue to use cooperative learning in working with my students to increase their achievement in problem solving and communication of mathematics.

Reading as a Learning Strategy for Mathematics

In this action research study of 55 sophomore and junior students in my Algebra II/Trigonometry classrooms, I investigated a reading strategy of learning mathematics. Students were given background information about reading and explored the benefits of reading for themselves. Next, students were taught to read their textbook, analyzing one section of the textbook at a time. Throughout the research project, students were given reading guides to fill out during class with whole class discussion following the reading time. I discovered that students are able to read a mathematics textbook with understanding and students who are gone for activities can learn independently. Teacher observations, student surveys, and student interviews provide quantitative evidence of increased student understanding and achievement. As a result of this research, I plan to continue utilizing the reading guides and incorporating reading as a method of learning mathematics within my classrooms.

Bandit-Based Algorithms for Budgeted Learning

We explore the problem of budgeted machine learning, in which the learning algorithm has free access to the training examples’ labels but has to pay for each attribute that is specified. This learning model is appropriate in many areas, including medical applications. We present new algorithms for choosing which attributes to purchase of which examples in the budgeted learning model based on algorithms for the multi-armed bandit problem. All of our approaches outperformed the current state of the art. Furthermore, we present a new means for selecting an example to purchase after the attribute is selected, instead of selecting an example uniformly at random, which is typically done. Our new example selection method improved performance of all the algorithms we tested, both ours and those in the literature.