Visualizing and Comparing Online Travel Reviews of the Great Walls: A Data Mining Approach

This research employs two samples of heritage sites of the Great Wall of China (Ba daling Great Wall and Mu tianyu Great Wall) and their 21000 reviews on TripAdvisor to visualize and induce feature-related comparisons. Word2vec and D3.js are applied for statistical computing and graphing Minimal Spanning Tree (MST) and ThemeRiver. The applications of MST and ThemeRiver are used to delineate outstanding features and clearer feature relationships. In terms of methodology, we applied an innovative research route to combine MST with ThemeRiver to visualize travellers’ online comments. At the same time, the visual results obtained are combined with qualitative analysis to generate valuable, intuitive summaries that can be used for reference in future research. Practically, the results disclose that although both sites are highly enjoyed by tourists, they are significantly different in terms of service, infrastructure and scenery. This article has implications for policymakers and practitioners with regard to making use of online reviews to gather authentic visitor comments on the Great Wall.

A Short-Cut Data Mining Method for the Mass Spectrometric Characterization of Block Copolymers

A new data mining approach as a short cut method is given for the determination of the copolymer composition from mass spectra. Our method simplifies the copolymer mass spectra by reduction of the number of mass peaks. The proposed procedure, namely the selection of the mass peaks, which is based on the most abundant peak of the mass spectrum, can be performed manually or more efficiently using our recently invented Mass-remainder analysis (MARA). The considerable reduction of the MS spectra also simplifies the calculation of the copolymer quantities for instance the number- and weight-average molecular weights (Mn and Mw, respectively), polydispersity index (Đ = Mw/Mn), average molar fraction (cA) and weight fraction (wA) of the comonomer A and so on. These copolymer properties are in line with those calculated by a reference method taking into account all the mass peaks of the copolymer distribution. We also suggest a highly efficient method and template for the determination of the composition drift by processing the reduced mass spectra.

Immunoinformatic Analysis Reveals Antigenic Heterogeneity of Epstein-Barr Virus Is Immune-Driven

Whole genome sequencing of Epstein-Barr virus (EBV) isolates from around the world has uncovered pervasive strain heterogeneity, but the forces driving strain diversification and the impact on immune recognition remained largely unknown. Using a data mining approach, we analyzed more than 300 T-cell epitopes in 168 published EBV strains. Polymorphisms were detected in approximately 65% of all CD8+ and 80% of all CD4+ T-cell epitopes and these numbers further increased when epitope flanking regions were included. Polymorphisms in CD8+ T-cell epitopes often involved MHC anchor residues and resulted in changes of the amino acid subgroup, suggesting that only a limited number of conserved T-cell epitopes may represent generic target antigens against different viral strains. Although considered the prototypic EBV strain, the rather low degree of overlap with most other viral strains implied that B95.8 may not represent the ideal reference strain for T-cell epitopes. Instead, a combinatorial library of consensus epitopes may provide better targets for diagnostic and therapeutic purposes when the infecting strain is unknown. Polymorphisms were significantly enriched in epitope versus non-epitope protein sequences, implicating immune selection in driving strain diversification. Remarkably, CD4+ T-cell epitopes in EBNA2, EBNA-LP, and the EBNA3 family appeared to be under negative selection pressure, hinting towards a beneficial role of immune responses against these latency type III antigens in virus biology. These findings validate this immunoinformatics approach for providing novel insight into immune targets and the intricate relationship of host defense and virus evolution that may also pertain to other pathogens.

A Wide Survey on Data Mining Approach for Crop Diseases Detection and Prevention

India is agriculture land and major revenue manufacturing sector. However, because of amendment in temporal parameters and uncertainty in climate directly have an effect on quality and amount of the assembly and maintenance of crops. Also, quality even a lot of degrade once the crops area unit infected by any malady. The main focus of this analysis in agriculture is to increment the crop quality and potency at lower price and gain profit as result of in India the majority of the population depends on agriculture. Big selection of fruits is growing up in India such as apple, banana, guava, grape, mango, pomegranate, orange is the main one. Fruit production gives around 20% of the country’s development. However, because of absence of maintenance, inappropriate development of fruits and manual investigation there has been scale back in generate the standard of fruits.So, Data Mining Approach used in the agriculture domain to resolve several agricultural issues of classification or prediction. During this paper complete survey of several data mining approach for crop disease management has been done. Detection of disease in early state will improve in quality of crop still as decrease the production cost. Also, we can improve the production of the particular crop. Several major parameters are used for the crop disease classification or prediction.