Normalização automática de descrições de hotéis

As descrições de produtos turísticos na área da hotelaria, aviação, rent-a-car e pacotes de férias baseiam-se sobretudo em descrições textuais em língua natural muito heterogénea com estilos, apresentações e conteúdos muito diferentes entre si. Uma vez que o sector do turismo é bastante dinâmico e que os seus produtos e ofertas estão constantemente em alteração, o tratamento manual de normalização de toda essa informação não é possível. Neste trabalho construiu-se um protótipo que permite a classificação e extracção automática de informação a partir de descrições de produtos de turismo. Inicialmente a informação é classificada quanto ao tipo. Seguidamente são extraídos os elementos relevantes de cada tipo e gerados objectos facilmente computáveis. Sobre os objectos extraídos, o protótipo com recurso a modelos de textos e imagens gera automaticamente descrições normalizadas e orientadas a um determinado mercado. Esta versatilidade permite um novo conjunto de serviços na promoção e venda dos produtos que seria impossível implementar com a informação original. Este protótipo, embora possa ser aplicado a outros domínios, foi avaliado na normalização da descrição de hotéis. As frases descritivas do hotel são classificadas consoante o seu tipo (Local, Serviços e/ou Equipamento) através de um algoritmo de aprendizagem automática que obtém valores médios de cobertura de 96% e precisão de 72%. A cobertura foi considerada a medida mais importante uma vez que a sua maximização permite que não se percam frases para processamentos posteriores. Este trabalho permitiu também a construção e população de uma base de dados de hotéis que possibilita a pesquisa de hotéis pelas suas características. Esta funcionalidade não seria possível utilizando os conteúdos originais. ABSTRACT: The description of tourism products, like hotel, aviation, rent-a-car and holiday packages, is strongly supported on natural language expressions. Due to the extent of tourism offers and considering the high dynamics in the tourism sector, manual data management is not a reliable or scalable solution. Offer descriptions - in the order of thousands - are structured in different ways, possibly comprising different languages, complementing and/or overlap one another. This work aims at creating a prototype for the automatic classification and extraction of relevant knowledge from tourism-related text expressions. Captured knowledge is represented in a normalized/standard format to enable new services based on this information in order to promote and sale tourism products that would be impossible to implement with the raw information. Although it could be applied to other areas, this prototype was evaluated in the normalization of hotel descriptions. Hotels descriptive sentences are classified according their type (Location, Services and/or Equipment) using a machine learning algorithm. The built setting obtained an average recall of 96% and precision of 72%. Recall considered the most important measure of performance since its maximization allows that sentences were not lost in further processes. As a side product a database of hotels was built and populated with search facilities on its characteristics. This ability would not be possible using the original contents.

Climate seasonality limits leaf carbon assimilation and wood productivity in tropical forests.

The seasonal climate drivers of the carbon cy- cle in tropical forests remain poorly known, although these forests account for more carbon assimilation and storage than any other terrestrial ecosystem. Based on a unique combina- tion of seasonal pan-tropical data sets from 89 experimental sites (68 include aboveground wood productivity measure- ments and 35 litter productivity measurements), their asso- ciated canopy photosynthetic capacity (enhanced vegetation index, EVI) and climate, we ask how carbon assimilation and aboveground allocation are related to climate seasonal- ity in tropical forests and how they interact in the seasonal carbon cycle. We found that canopy photosynthetic capacity seasonality responds positively to precipitation when rain- fall is < 2000 mm yr-1 (water-limited forests) and to radia- tion otherwise (light-limited forests). On the other hand, in- dependent of climate limitations, wood productivity and lit- terfall are driven by seasonal variation in precipitation and evapotranspiration, respectively. Consequently, light-limited forests present an asynchronism between canopy photosyn- thetic capacity and wood productivity. First-order control by precipitation likely indicates a decrease in tropical forest pro- ductivity in a drier climate in water-limited forest, and in cur- rent light-limited forest with future rainfall < 2000 mm yr-1.

Energy balances in sugar cane, coffee and natural vegetation in the northeastern side of the São Paulo state, Brazil

Under land and climate change scenarios, agriculture has experienced water competitions among other sectors in the São Paulo state, Brazil. On the one hand, in several occasions, in the northeastern side of this state, nowadays sugar-cane is expanding, while coffee plantations are losing space. On the other hand, both crops have replaced the natural vegetation composed by Savannah and Atlantic Coastal Forest species. Under this dynamic situation, geosciences are valuable tools for evaluating the large-scale energy and mass exchanges between these diffe rent agro-ecosystems and the lower atmosphere. For quantification of the energy balance components in these mixed agro-ecosystems, the bands 1 and 2 from the MODIS product MOD13Q1 we re used throughout SA FER (Surface Algorithm for Evapotranspiration Retrieving) algorithm, which was applied together with a net of 12 automatic weather stations, during the year 2015 in the main sugar cane and coffee growing regions, located at the no rtheastern side of the state. The fraction of the global solar radiation (R G ) transformed into net radiation (Rn) was 52% for sugar cane and 53% for both, coffee and natural vegetation. The respective annual fractions of Rn used as λ E were 0.68, 0.87 and 0.77, while for the sensible heat (H) fluxes they were 0.27, 0.07 and 0.16. From April to July, heat advection raised λ E values above Rn promoting negative H, however these effects were much and less strong in coffee and sugar cane crop s, respectively. The smallest daily Rn fraction for all agro-ecosystems was for the soil heat flux (G), with averages of 5%, 6% and 7% in sugar cane, coffee and natural vegetation. From the energy balance analyses, we could conclude that, sugar-cane crop presented lower annual water consumption than that for coffee crop , what can be seen as an advantage in situations of water scarcity. However, the replacement of natural vegetation by su gar cane can contribute for warming th e environment, while when this occur with coffee crop there was noticed co oling conditions. The large scale modeling satisfactory results confirm the suitability of using MODIS products togeth er with weather stations to study the energy balance components in mixed agro-ecosystems under land-use and climate change conditions.