Database Management Research Proposal of Wine Quality

Another round?

Database Management Research Proposal of Wine Quality

Project Scope:

Our initiative is to build up a database of physicochemical property of a list of wines. Users could easily add, edit or view the acidity, PH level, or residual sugar level of a particular wine. Based on information we have, we would also be doing predictive data analysis and classification analysis. This would allow us to know which set of factors would mostly have strong influence on wine quality, which set would be relatively “weak”; whether this influence would be positive or negative; what physical properties would make a “perfect” wine.

http://archive.ics.uci.edu/ml/datasets/Wine+Quality

Our project data would be sourced from UCI machine learning portal. This is a multivariate dataset containing output sensory variable of Portuguese "Vinho Verde" wine such as quality (ranking from 0 to 10) and input physicochemical input variables such as 1 - fixed acidity 2 - volatile acidity 3 - citric acid 4 - residual sugar 5 - chlorides 6 - free sulfur dioxide 7 - total sulfur dioxide 8 - density 9 - pH 10 - sulphates 11 - alcohol.

The reason why we want to study this topic arises in several aspects. First, certification and quality assessment are crucial issues within the wine industry. Wine industry demonstrates a late development spurt as social drinking is on the ascent. A glass of wine a day keeps the doctor away. Furthermore, wine quality reports are produced annually, but relevant scientific data on wines are outdated, need to implement database to better organize wine related information. And also quality assessment is frequently part of the confirmation procedure and can be utilized to enhance wine making (by recognizing the most compelling components) and to stratify wines, for example, premium brands (helpful at setting costs).

This study is valuable not only for certification entities but also wine producers and even consumers. It can be used to support the oenologist's wine evaluations, potentially improving the quality and speed of their decisions. In addition, measuring the impact of the physicochemical tests in the final wine quality is useful for improving the production process. Furthermore, it can help in target marketing, i.e. by applying similar techniques to model the consumer's preferences of niche and/or profitable markets. For wine tasters/ consumers, help them understand what influence their taste most.

Literature Review

1.)Scientist has been focusing on what could make the best wine as early as 1980s. In this paper “Influence of cluster exposure on fruit composition and wine quality of Seyval blanc grapes”. Experiments were conducted in 1981 and 1982 to investigate the influence of duster exposure on fruit composition, wine quality, and incidence of bunch rot (Botrytis cinerea PERS.) of the French-American hybrid grape cultivar Seyval blanc. Berry sampling from 4 exposure categories indicated highest Brix, pH and tartrate, and lowest titratable acidity and malate in the exposed fruit post veraison. Total acidity and malate were highest in exposed fruit between fruit set and veraison. Exposure of the fruit also reduced the incidence of bunch rot during the 1981 season. Wine quality differences were small and not statistically significant, although wines vinted from exposed fruit tended to score higher. Canopy management practices that optimize fruit exposure would be helpful in maximization of fruit and wine quality. But however, the data behind is too old and needed to be updated for modern consumer taste.

2.)In “Climate Change and Global Wine Quality”, Gregory V. Jones et al 2005 found out that with an averaging growing in temperature 2 Celsius, climate changes have some impact statistically significant related to wine quality in some regions. This study reveals that the climate changes impact for wine grape growing are not likely to be uniform across regions. As a result, some high quality regions will reach the margins of their climatic limits, while some regions will be pushed to more optimal climatic regimes. But this paper didn’t used database to optimize the storage of that data, it also didn’t relate the climate change effects with physical characteristics of the wine data.

3.) In another article “Using data mining techniques to predict industrial wine

problem fermentations”, winemakers at present do not have the tools to distinguish early indications of fermentation aging conducted and are unable to mitigate it. Information gathered from following 24 industrial fermentation of Cabernet sauvignon were utilized as a part of this study to investigate how valuable is information mining to distinguish bizarre practices ahead of time. A database held occasional estimations of 29 segments that included sugar, alcohols, natural acids and amino acids. Attributable to the size of the issue, we utilized a two-stage classification method. Initially PCA was utilized to lessen framework dimensionality while safeguarding metabolite cooperation data. Cluster examination (K-Means) was then performed on the lower-dimensioned framework to gathering fermentations into groups of comparative conduct. Various groupings were investigated relying upon the information utilized. At first information from simply the initial three days were evaluated, and after that the whole information set was utilized. Data from the initial three days' fermentation conduct gives essential insights about the last order. We additionally found a solid relationship between problematic fermentations and particular patterns found by the information mining apparatuses. To put it plainly, information from the initial three days contain adequate data to set up the probability of an aging completing typically. Comes about because of this study are generally reassuring. Information from numerous more fermentations and of various assortments should be gathered, be that as it may, to build up a solid and all the more extensively appropriate indicative device.

4.)Apart from biology part of the story, economist has also been studying how wines prices and wine qualities relate to each other. This paper “Wine Quality and Varietal, Regional and Winery Reputations: Hedonic Prices for Australia and New Zealand” by Günter Schamel,

and Kym Anderson. They estimate hedonic price functions for premium wine from Australia and New Zealand, differentiating implicit prices for sensory quality ratings, wine varieties and regional as well as winery brand reputations over the vintages 1992–2000. The results show regional reputations have become increasingly differentiated through time (although less so for New Zealand). In particular, cool-climate regions are becoming increasingly preferred over other regions in Australia. In each country, price premia associated with both James Halliday's and Winestate magazine's sensory quality ratings, and with Halliday's winery ratings and classic wine designations, are highly significant.