Uma abordagem baseada em filtragem colaborativa integrada a mapas de saliência para a recomendação de imagens

Nowadays, the amount of customers using sites for shopping is greatly increasing, mainly due to the easiness and rapidity of this way of consumption. The sites, differently from physical stores, can make anything available to customers. In this context, Recommender Systems (RS) have become indispensable to help consumers to find products that may possibly pleasant or be useful to them. These systems often use techniques of Collaborating Filtering (CF), whose main underlying idea is that products are recommended to a given user based on purchase information and evaluations of past, by a group of users similar to the user who is requesting recommendation. One of the main challenges faced by such a technique is the need of the user to provide some information about her preferences on products in order to get further recommendations from the system. When there are items that do not have ratings or that possess quite few ratings available, the recommender system performs poorly. This problem is known as new item cold-start. In this paper, we propose to investigate in what extent information on visual attention can help to produce more accurate recommendation models. We present a new CF strategy, called IKB-MS, that uses visual attention to characterize images and alleviate the new item cold-start problem. In order to validate this strategy, we created a clothing image database and we use three algorithms well known for the extraction of visual attention these images. An extensive set of experiments shows that our approach is efficient and outperforms state-of-the-art CF RS.

Frequência de microrganismos em conteúdo vulvovaginal de vacas tratadas com dispositivos intravaginais de progesterona

The dissertation was divided in two studies. With the first, aimed to evaluate the occurrence of microorganisms present in the vulvovaginal region of cows that received intravaginal progesterone devices during the fixed-time artificial insemination (FTAI) programs, and correlate the results with pregnancy rates. Samples were collected from vulvovaginal region of 30 beef cows Guzerá and 30 crossbred dairy cows, and also intravaginal devices, randomly. Of the 120 samples of cows, 60 corresponded to the collections of the period prior to the introduction device (D0) and 60 to the subsequent withdrawal of it (D9); it yielded 100% of bacterial growth, whereas, in most samples, it was found more than one isolated. In D0, the most frequent agent was Escherichia coli (52%), and in D9, Proteus spp and E. coli were the most frequent (32% and 28%, respectively). Regarding intravaginal progesterone devices, in D0 were isolated 37 microorganisms, being predominant those of the genus Bacillus (35%); in D9, 41 colony forming units (CFU) were isolated, of which 36.6% corresponded to Proteus spp. For the analysis of the antimicrobial profile, susceptibility testing was performed by diffusion agar disk, and cows that did not became pregnant after FTAI program were selected, as a future treatment. There resistance 100% to penicillin, and sensitivity, approximately, 90% to gentamicin, both isolates obtained from samples of beef cows and obtained of dairy cows. Regarding pregnancy rate, the 30 beef cows, 11 were diagnosed pregnant (36.7%), 4 (36.4%) treated with reused devices and 7 (63.6%) with new devices, which showed more effective. Of the 30 dairy cows, 15 were pregnant (50%), 8 (53.3%) were implanted with reused devices and 7 (46.7%) with new devices, with no significant differences in pregnancy rates. Because it is a research, females were chosen at random, and factors such as body condition, nutritional management and health weren't priority. With the second study, aimed to analyze the similarity between strains, conducted by technical Random Amplified Polymorphic DNA (RAPD -PCR). Presence of E. coli and the absence of pregnancy were selection criteria used. From the results, it was observed that most of the isolates wasn't phylogenetically similar, since they showed lower than 85% similarity. The study stressed the importance of E. coli in vulvovaginal microbiota of cows and the presence of phenotypic and genotypic characters of this bacterium on possible reproductive problems.