Estudo comparativo entre técnicas de aprendizado de máquina para estimação de risco de crédito

Dissertação (mestrado)—Universidade de Brasília, Faculdade de Economia, Administração e Contabilidade, Programa de Pós-Graduação em Administração, 2016.

COMMUNITY KNOWLEDGE, ATTITUDE AND PRACTICES (KAP) ON FLUOROSIS AND ITS MITIGATION IN ENDEMIC AREAS OF ETHIOPIA

Optimum fluoride intake plays an essential role in the prevention of dental caries while fluoride consumption above recommended level interferes with the normal formation of tooth enamel and bones and may increase risk of dental and skeletal fluorosis. The knowledge and practices of endemic communities on etiology of fluorosis will help in its mitigation and prevention. The objective of this study was to investigate the knowledge, attitude and practices of endemic community on fluoride contamination, fluorosis and prevention practices in order to devise coordinated and targeted prevention mechanisms. Focus group discussions (FGD) and key-informant interview were conducted in three dietary areas to collect knowledge, attitude and practices (KAP) of the endemic community in July 2013.The results indicated that health consequences of fluoride contaminated water are fairly understood. None of the discussants mentioned the word “fluoride”. The knowledge and perception of the community on fluoride ingestion is poor. Health extension workers (HEWs) did not teach about fluoride and related health consequences. Dental fluorosis was reported to start at early ages and not commonly perceived as a major problem. However, adolescents worried and felt that they might be singled out when going to other areas. Older people have a skeletal fluorosis, which interferes with their day to day activities. In severely affected people, the teeth were weak and fragile and thus create difficulty in chewing hard foods like unfermented dry flat bread, sugar cane and toasted grains. People prefer rain water rather than water from borehole because of the inconvenient taste of the latter. The endemic communities have no sufficient knowledge and skills on potential sources of fluoride intake, the debilitating effect of high fluoride ingestion, and preventive and mitigatory measures to reduce fluoride intake. The effect of fluoride contamination and mitigatory methods should get sufficient attention by the community, health workers and concerned governmental bodies. The trend of harvesting and using rain water should be encouraged as it reduces fluoride intake. Future studies should focus on information communication on possible fluoride risks, intervention and evaluation studies on defluoridation, rain water harvesting and mitigatory techniques.

Healthcare providers’ knowledge, attitude and behaviour towards breast cancer diagnosis and treatment in Malaysia – a mini systematic review

Breast cancer is the most common cancer among women in Malaysia. Therefore, it is important for the public to be educated on breast cancer and to know the steps that need to be taken to detect it early. Healthcare providers are in a unique position to provide public health education due to their good knowledge of health issues and their roles in healthcare. A systematic review of studies conducted from 2008 till 2015 was undertaken to analyze the knowledge, attitudes and behavior of Malaysian healthcare providers regarding breast cancer, in an attempt to obtain an overall picture of how wellequipped the healthcare providers are to provide optimal breast cancer education, and to ascertain their perceptions and actual involvement in such education. The systematic review was conducted via a primary search of various databases and journal websites, and a secondary search of references cited in eligible studies. Criteria for eligibility include studies conducted in Malaysia and published from the year 2008 to 2015, and written in English language. A total of fifteen articles were identified and reviewed but only two studies were eligible for this review. The findings suggest that future and current Malaysian healthcare providers have moderate knowledge of breast cancer, showed a positive disposition towards involvement in breast cancer education, but displayed poor involvement.

Public and private initiatives for regional development : an investigation of knowledge transfer and social entrepreneurship

Otto-von-Guericke-Universität Magdeburg, Fakultät für Wirtschaftswissenschaft, Dissertation, 2016

Collaborative filtering and deep learning based recommendation system for cold start items

Recommender system is a specific type of intelligent systems, which exploits historical user ratings on items and/or auxiliary information to make recommendations on items to the users. It plays a critical role in a wide range of online shopping, e-commercial services and social networking applications. Collaborative filtering (CF) is the most popular approaches used for recommender systems, but it suffers from complete cold start (CCS) problem where no rating record are available and incomplete cold start (ICS) problem where only a small number of rating records are available for some new items or users in the system. In this paper, we propose two recommendation models to solve the CCS and ICS problems for new items, which are based on a framework of tightly coupled CF approach and deep learning neural network. A specific deep neural network SADE is used to extract the content features of the items. The state of the art CF model, timeSVD++, which models and utilizes temporal dynamics of user preferences and item features, is modified to take the content features into prediction of ratings for cold start items. Extensive experiments on a large Netflix rating dataset of movies are performed, which show that our proposed recommendation models largely outperform the baseline models for rating prediction of cold start items. The two proposed recommendation models are also evaluated and compared on ICS items, and a flexible scheme of model retraining and switching is proposed to deal with the transition of items from cold start to non-cold start status. The experiment results on Netflix movie recommendation show the tight coupling of CF approach and deep learning neural network is feasible and very effective for cold start item recommendation. The design is general and can be applied to many other recommender systems for online shopping and social networking applications. The solution of cold start item problem can largely improve user experience and trust of recommender systems, and effectively promote cold start items.

Firms’ knowledge search and local knowledge externalities in innovation performance

We use an augmented version of the UK Innovation Surveys 4–7 to explore firm-level and local area openness externalities on firms’ innovation performance. We find strong evidence of the value of external knowledge acquisition both through interactive collaboration and non-interactive contacts such as demonstration effects, copying or reverse engineering. Levels of knowledge search activity remain well below the private optimum, however, due perhaps to informational market failures. We also find strong positive externalities of openness resulting from the intensity of local interactive knowledge search—a knowledge diffusion effect. However, there are strong negative externalities resulting from the intensity of local non-interactive knowledge search—a competition effect. Our results provide support for local initiatives to support innovation partnering and counter illegal copying or counterfeiting. We find no significant relationship between either local labour quality or employment composition and innovative outputs.

Collaborative filtering and deep learning based hybrid recommendation for cold start problem

Recommender systems (RS) are used by many social networking applications and online e-commercial services. Collaborative filtering (CF) is one of the most popular approaches used for RS. However traditional CF approach suffers from sparsity and cold start problems. In this paper, we propose a hybrid recommendation model to address the cold start problem, which explores the item content features learned from a deep learning neural network and applies them to the timeSVD++ CF model. Extensive experiments are run on a large Netflix rating dataset for movies. Experiment results show that the proposed hybrid recommendation model provides a good prediction for cold start items, and performs better than four existing recommendation models for rating of non-cold start items.

Knowledge of and attitudes toward sexual and reproductive health among college students

Abstract Introduction: Knowledge provides the foundation for values, attitudes and behavior. Knowledge about sexual and reproductive health (SRH) and positive attitudes are essential for implementing protective behaviors. Objectives: The aim of this study was to evaluate SRH knowledge and attitudes in college students and their association with sexual and reproductive behaviors. Material and methods: A cross-sectional study was conducted in a sample of 1946 college students. The data were collected using a self-report questionnaire on the sociodemographics characteristics of the sample, an inventory on SRH knowledge and an attitude scale, and were analyzed with descriptive and inferential statistics (ANOVA and Pearson’s correlation). Results: The sample was 64% female and 36% male, with a mean age of 21 years. The majority were sexually active and used contraception. The SRH knowledge was moderate (22.27 ± 5.79; maximum score = 44), while the average SRH attitude score was more favorable (118.29 ± 13.92; maximum score = 140). Female and younger students studying life and health sciences had higher (P < .05) SRH knowledge and attitude scores. The consistent use of condom and health care surveillance were highly dependent on the students’ SRH knowledge and attitudes. Engagement in sexual risk behaviors was associated with lower scores for these variables. Conclusions: Strategies to increase SRH knowledge and attitudes are important tools for improving protective behaviors, especially with respect to contraception, health care surveillance and exposure to sexual risk. Older males studying topics other than life sciences should be a priority target for interventions due to their higher sexual risk

Subject Ontogeny: Subject Access through Time and the Dimensionality of Classification

Classification schemes undergo revision. However, in a networked environment revisions can be used to add dimensionality to classification. This dimensionality can be used to help explain conceptual warrant, explain the shift from disciplinary to multidisciplinary knowledge production, and as a component method of domain analysis. Further, subject ontogeny might be used in cooperative networked projects like digital preservation, online access tools, and interoperability frameworks.

Using online peer assessment tools to provide feedback and promote learning

A set of slides used for the RAP SIG event on 19 Jan 2017

Analysis of Dyscalculia Evidences through Artificial Intelligence Systems

Dyscalculia is usually perceived of as a specific learning difficulty for mathematics or, more appropriately, arithmetic. Because definitions and diagnoses of dyscalculia are in their infancy and sometimes are contradictory. However, mathematical learning difficulties are certainly not in their infancy and are very prevalent and often devastating in their impact. Co-occurrence of learning disorders appears to be the rule rather than the exception. Co-occurrence is generally assumed to be a consequence of risk factors that are shared between disorders, for example, working memory. However, it should not be assumed that all dyslexics have problems with mathematics, although the percentage may be very high, or that all dyscalculics have problems with reading and writing. Because mathematics is very developmental, any insecurity or uncertainty in early topics will impact on later topics, hence to need to take intervention back to basics. However, it may be worked out in order to decrease its degree of severity. For example, disMAT, an app developed for android may help children to apply mathematical concepts, without much effort, that is turning in itself, a promising tool to dyscalculia treatment. Thus, this work will focus on the development of a Decision Support System to estimate children evidences of dyscalculia, based on data obtained on-the-fly with disMAT. The computational framework is built on top of a Logic Programming approach to Knowledge Representation and Reasoning, grounded on a Case-based approach to computing, that allows for the handling of incomplete, unknown, or even self-contradictory information.

Knowledge, perceptions and attitudes on the sexuality of nursing students and teachers

Introduction: The personal attitudes regarding specific aspects of sexuality are of interest to practices of personal concern, as they are to practices inserted in professional roles. General attitudes towards sexuality and sexual health were evaluated. Objectives: To describe the perceptions and attitudes of students and nursing teachers about sexuality. Methods: We used a mixed methods design with a sequential strategy: QUAN→qual of descriptive and explanatory type. 646 students and teachers participated. The Sexual Attitudes Scale (EAS) of Hendrick & Hendrick (Alferes, 1999) and Attitude Scale Address Sexual and Reproductive Health (EAFSSR) of Nemčić et al (Abreu, 2008) were used. Results: There are significant differences in the level of knowledge about sexuality depending on the sample (χ2KW (2)=18.271; p=.000): students of 1st year have lower levels. The profile of the four dimensions of EAS per sample is identical in all 3 samples, having responsibility the highest average value. In subscales EAFSSR per sample and sex there are significant diferences (p<.05) for all samples and uniform pattern was noted: females have higher median values, indicating that they have more favorable attitudes towards sexual health. Conclusions: Sexual attitudes reveal a multidimensional structure based in the female identity, that shows responsibility towards family planning and sexual education, as well as towards individual self-care regarding the body and sexual and reproductive health. An attitudinal profile by gender emerges, accentuating the polarity between male and female. The importance of the training process in nursing following the personal and social development of students is corroborated.

Tecniche di machine learning per type inference di linguaggi tipati dinamicamente

In questa tesi vengono discusse le principali tecniche di machine learning riguardanti l'inferenza di tipo nei linguaggi tipati dinamicamente come Python. In aggiunta è stato creato un dataset di progetti Python per l'addestramento di modelli capaci di analizzare il codice

Rilevamento della qualità del manto stradale: un approccio sperimentale tramite Machine Learning

Il riconoscimento delle condizioni del manto stradale partendo esclusivamente dai dati raccolti dallo smartphone di un ciclista a bordo del suo mezzo è un ambito di ricerca finora poco esplorato. Per lo sviluppo di questa tesi è stata sviluppata un'apposita applicazione, che combinata a script Python permette di riconoscere differenti tipologie di asfalto. L’applicazione raccoglie i dati rilevati dai sensori di movimento integrati nello smartphone, che registra i movimenti mentre il ciclista è alla guida del suo mezzo. Lo smartphone è fissato in un apposito holder fissato sul manubrio della bicicletta e registra i dati provenienti da giroscopio, accelerometro e magnetometro. I dati sono memorizzati su file CSV, che sono elaborati fino ad ottenere un unico DataSet contenente tutti i dati raccolti con le features estratte mediante appositi script Python. A ogni record sarà assegnato un cluster deciso in base ai risultati prodotti da K-means, risultati utilizzati in seguito per allenare algoritmi Supervised. Lo scopo degli algoritmi è riconoscere la tipologia di manto stradale partendo da questi dati. Per l’allenamento, il DataSet è stato diviso in due parti: il training set dal quale gli algoritmi imparano a classificare i dati e il test set sul quale gli algoritmi applicano ciò che hanno imparato per dare in output la classificazione che ritengono idonea. Confrontando le previsioni degli algoritmi con quello che i dati effettivamente rappresentano si ottiene la misura dell’accuratezza dell’algoritmo.