Study And Prediction Of Quality Of Wine Based On Pattern Recognition

With the rapid development of society, more and more food and wine are enjoyed bymore and more people. Once viewed as a luxury good, nowadays wine is increasingly enjoyedby a wider range of consumers. Now not only in wide world, but also in China, the market ofwine is increasing rapidly, and more and more wine industry are focusing on the Chinamarket. To support its growth, the wine industry is investing in new technologies for bothwine making and selling processes. Wine certification and quality assessment are keyelements within this context. Certification prevents the illegal adulteration of wines (tosafeguard human health) and assures quality for the wine market.Quality evaluation is often part of the certification process and can be used to improvewine making (by identifying the most influential factors) and to stratify wines such aspremium brands (useful for setting prices). Wine certification is generally assessed byphysicochemical and sensory tests. Physicochemical laboratory tests routinely used tocharacterize wine include determination of density, alcohol or pH values, while sensory testsreply mainly on human experts. It should be stressed that taste is the least understood of thehuman senses, thus wine classification is a difficult task. Moreover, the relationships betweenthe physicochemical and sensory analysis are complex and still not fully understood.Advances in information technologies have made it possible to collect, store and processmassive, often highly complex datasets. All this data hold valuable information such as trendsand patterns, which can be used to improve decision making and optimize chances of success.This paper first introduces the history and development of the Chinese wine a nd winestatus, and point out current problem. Because it’s too complex during wine production, we suppose it can be solved by the method of pattern recognition.Since1960of the20th century, the research has made great progress on theory andmethod of pattern recognition and its application in engineering.Pattern recognition is generally categorized according to the type of learning procedureused to generate the output value. Supervised learning assumes that a set of training data (thetraining set) has been provided, consisting of a set of instances that have been properlylabeled by hand with the correct output. A learning procedure then generates a model thatattempts to meet two sometimes conflicting objectives: Perform as well as possible on thetraining data, and generalize as well as possible to new data. Unsupervised learning, on theother hand, assumes training data that has not been hand-labeled, and attempts to find inherentpatterns in the data that can then be used to determine the correct output value for new datainstances.We proposed pattern recognition approach be used to analyze and predict the quality ofwine. And the effectiveness of using pattern recognition approach has been proved byexperiments. In order to solve the slow problem of supervision (classification) algorithms, weproposed combining with non-supervision (clustering) algorithm to increase the speed ofclassification and prediction. By analyzing the experimental data, the approach of combiningthem together to speed up the prediction is valid, and we also point out limitation about thecombined algorithm.