Preparing the books to read every day is enjoyable for many people. However, there are still many people who also don't like reading. This is a problem. But, when you can support others to start reading, it will be better. One of the books that can be recommended for new readers is experience and education. This book is not kind of difficult book to read. It can be read and understand by the new readers.

Experience and Education.

The use of recommender systems is an emerging trend today, when user behavior information is abundant. There are many large datasets available for analysis because many businesses are interested in future user opinions. Sophisticated algorithms that predict such opinions can simplify decision-making, improve customer satisfaction, and increase sales. However, modern datasets contain millions of records, which represent only a small fraction of all possible data. Furthermore, much of the information in such sparse datasets may be considered irrelevant for making individual recommendations. As a result, there is a demand for a way to make personalized suggestions from large amounts of noisy data.
Current recommender systems are usually all-in-one applications that provide one type of recommendation. Their inflexible architectures prevent detailed examination of recommendation accuracy and its causes. We introduce a novel architecture model that supports scalable, distributed suggestions from multiple independent nodes. Our model consists of two components, the input matrix generation algorithm and multiple platform-independent combination algorithms. A dedicated input generation component provides the necessary data for combination algorithms, reduces their size, and eliminates redundant data processing. Likewise, simple combination algorithms can produce recommendations from the same input, so we can more easily distinguish between the benefits of a particular combination algorithm and the quality of the data it receives. Such flexible architecture is more conducive for a comprehensive examination of our system.
We believe that a user's future opinion may be inferred from a small amount of data, provided that this data is most relevant. We propose a novel algorithm that generates a more optimal recommender input. Unlike existing approaches, our method sorts the relevant data twice. Doing this is slower, but the quality of the resulting input is considerably better. Furthermore, the modular nature of our approach may improve its performance, especially in the cloud computing context. We implement and validate our proposed model via mathematical modeling, by appealing to statistical theories, and through extensive experiments, data analysis, and empirical studies.
Our empirical study examines the effectiveness of accuracy improvement techniques for collaborative filtering recommender systems. We evaluate our proposed architecture model on the Netflix dataset, a popular (over 130,000 solutions), large (over 100,000,000 records), and extremely sparse (1.1%) collection of movie ratings. The results show that combination algorithm tuning has little effect on recommendation accuracy. However, all algorithms produce better results when supplied with a more relevant input. Our input generation algorithm is the reason for a considerable accuracy improvement.

/pdf/a-distributed-architecture-centric-approach-to-computing-31zmh6iinx.pdf

A distributed, architecture-centric approach to computing accurate recommendations from very large and sparse datasets

Trabajo presentado en el 39th CIESM Congress, celebrado en Venecia, Italia, del 10 al 14 de mayo de 2010

/pdf/biodiversity-of-the-mediterranean-sea-estimates-patterns-and-2mjs6mi5cx.pdf

Biodiversity of the Mediterranean Sea: estimates, patterns and threats

In line with global targets agreed under the Convention on Biological Diversity, the number of marine protected areas (MPAs) is increasing rapidly, yet socio-economic benefits generated by MPAs remain difficult to predict and under debate1, 2. MPAs often fail to reach their full potential as a consequence of factors such as illegal harvesting, regulations that legally allow detrimental harvesting, or emigration of animals outside boundaries because of continuous habitat or inadequate size of reserve3, 4, 5. Here we show that the conservation benefits of 87 MPAs investigated worldwide increase exponentially with the accumulation of five key features: no take, well enforced, old (>10 years), large (>100 km2), and isolated by deep water or sand. Using effective MPAs with four or five key features as an unfished standard, comparisons of underwater survey data from effective MPAs with predictions based on survey data from fished coasts indicate that total fish biomass has declined about two-thirds from historical baselines as a result of fishing. Effective MPAs also had twice as many large (>250 mm total length) fish species per transect, five times more large fish biomass, and fourteen times more shark biomass than fished areas. Most (59%) of the MPAs studied had only one or two key features and were not ecologically distinguishable from fished sites. Our results show that global conservation targets based on area alone will not optimize protection of marine biodiversity. More emphasis is needed on better MPA design, durable management and compliance to ensure that MPAs achieve their desired conservation value.

/pdf/global-conservation-outcomes-depend-on-marine-protected-2bjbqmu98v.pdf

Global conservation outcomes depend on marine protected areas with five key features

The Mediterranean Sea is a marine biodiversity hot spot. Here we combined an extensive literature analysis with expert opinions to update publicly available estimates of major taxa in this marine ecosystem and to revise and update several species lists. We also assessed overall spatial and temporal patterns of species diversity and identified major changes and threats. Our results listed approximately 17,000 marine species occurring in the Mediterranean Sea. However, our estimates of marine diversity are still incomplete as yet—undescribed species will be added in the future. Diversity for microbes is substantially underestimated, and the deep-sea areas and portions of the southern and eastern region are still poorly known. In addition, the invasion of alien species is a crucial factor that will continue to change the biodiversity of the Mediterranean, mainly in its eastern basin that can spread rapidly northwards and westwards due to the warming of the Mediterranean Sea. Spatial patterns showed a general decrease in biodiversity from northwestern to southeastern regions following a gradient of production, with some exceptions and caution due to gaps in our knowledge of the biota along the southern and eastern rims. Biodiversity was also generally higher in coastal areas and continental shelves, and decreases with depth. Temporal trends indicated that overexploitation and habitat loss have been the main human drivers of historical changes in biodiversity. At present, habitat loss and degradation, followed by fishing impacts, pollution, climate change, eutrophication, and the establishment of alien species are the most important threats and affect the greatest number of taxonomic groups. All these impacts are expected to grow in importance in the future, especially climate change and habitat degradation. The spatial identification of hot spots highlighted the ecological importance of most of the western Mediterranean shelves (and in particular, the Strait of Gibraltar and the adjacent Alboran Sea), western African coast, the Adriatic, and the Aegean Sea, which show high concentrations of endangered, threatened, or vulnerable species. The Levantine Basin, severely impacted by the invasion of species, is endangered as well.

This abstract has been translated to other languages (File S1).

/pdf/the-biodiversity-of-the-mediterranean-sea-estimates-patterns-38y4s7jond.pdf

The Biodiversity of the Mediterranean Sea: Estimates, Patterns, and Threats

It is with particular pride that the Italian Journal of Zoology opens this volume with an invited review on eye evolution referring to Darwin in the title. The senior author, Walter Gehring, labels himself a zoologist (having also authored a zoology textbook) and uses molecular approaches just as a key to open doors that have been identified by classical zoology, starting from Darwin’s times. If looked at from a morphological point of view, eyes give the impression of having been invented several times during animal evolution. The first animals with true eyes are the jellyfish (with cubozoan eyes having even a lens, a cornea and a retina, just like our eyes). Throughout the metazoa, eyes evolved so to acquire much different structures, all performing the same function: get information from the outer world through light reception. Did they evolve independently, and so are polyphyletic, or did they evolve just once, to be then re-elaborated by evolution so to reach many different levels of organization, being nonetheless monophyletic? Morphology and developmental biology hint that the origin is polyphyletic (Salvini-Plawen and Mayr 1977) but genetics show that the genes that switch on the process of eye formation are very conserved, and probably evolved just once (Gehring and Seimiya 2010). Vertebrate wings are a classical controversial case of analogy vs. homology. The wings of extinct reptiles, birds, and mammals are obviously polyphyletic as wings, but are monophyletic as fore-limbs of tetrapods. The origin is common, then evolution made them reach different levels of organization while converging to the same function. In a way, the advocates of monophyly and those of polyphyly are both right. Long time ago, this Journal started a debate about the Cnidaria representing the premises of metazoan evolution (Boero et al. 1998). Since then, more and more evidence has been accumulated, showing that the evolutionary innovations present in the cnidaria, the most ancient of the living metazoa, have been inherited by the rest of the metazoan clade, and that what can be considered as a grade of organization might hide a clade. Evolution did not proceed in a linear way, though, and this is being revealed by the fact that different clades can often be obtained while considering the same set of species, if different characters are considered, both at molecular or at morphological levels. It is not matter of who is right and who is wrong, here. Evolution is both saltational and gradual, some features evolve very quickly, others remain very conserved. There is no universal law in evolution, besides the fact that evolution takes place. Walter Gehring and Makiko Seimiya elegantly show that evolution is both innovative and conservative, recognizing the fundamental role of observation and comparison, the main tools utilized by Darwin, to generate hypotheses that are then tested with all the tools that technology makes available. These tools, however, would be as useless as a key without a door to open, if observation and comparison were no longer performed.

An eye on evolution

Morphogenesis and developmental biology

The Role of Local to Large‐Scale Connectivity in Marine Community Recovery Dynamics

Current Trends in Hydrozoan Biology - VI

Previously, the Italian Journal of Zoology was divided into sections, covering the manifold approaches of zoology. Sections, however, were often unbalanced, with rather few contributions based eith...

Ferdinando Boero

Papers

An eye on evolution

Morphogenesis and developmental biology

The Role of Local to Large‐Scale Connectivity in Marine Community Recovery Dynamics

Current Trends in Hydrozoan Biology - VI

The many faces of zoology