Examples of data mining

Agriculture

• Data mining techniques can be applied to visual data in agriculture to extract meaningful patterns, trends, and associations. This information can improve algorithms that detect defects in harvested fruits and vegetables. For example, advanced visual data collection methods, machine vision systems, and image processing have been applied to classify fruits and vegetables according to numerous surface defects. Additionally, this data can be analyzed to investigate potential causes of defects. Much of this knowledge is based on anecdotal evidence rather than qualitative and quantitative data collection methods. However, efforts are being made to integrate data mining techniques into horticulture research. Before being sent to market, apples are inspected, and those with defects are removed. However, invisible defects can affect an apple's flavor and appearance. One example of such is the water core, an internal disorder that can affect the fruit's shelf life. Apples with a slight water core are sweeter, but those with a moderate to severe water core have reduced storage potential compared to regular apples. Additionally, a few fruits with severe water cores could spoil an entire batch. Because of this, a computational system is under study that captures X-ray images of the apples as they run on conveyor belts. The system analyzes the pictures using machine learning algorithms to predict the likelihood of the fruit containing water cores.
• The metabolic transformations during fermentation impact the quality of the wine produced and the productivity of wine production industries. Data science techniques, such as k-means clustering, and classification techniques based on biclustering, have been used to study these metabolic processes, successfully predicting fermentation outcomes after as little as three days. These methods allow for the classification of wine according to the metabolite profile of the fermentation, differentiating it from traditional wine classification systems.
• A Group Method of Data Handling -type network, combined with a genetic algorithm, was used to predict the metabolizable energy of feather meal and poultry offal meal based on protein, fat, and ash content. Data samples from published literature were collected and used to train a GMDH-type network model. This approach can predict the metabolizable energy of poultry feed samples based on their chemical content. The GMDH-type network can also accurately estimate poultry performance from dietary nutrients such as metabolizable energy, protein and amino acids.
• The early detection of animal diseases can benefit farm productivity by allowing farmers to treat and isolate affected animals as soon as symptoms appear, reducing the spread of disease. For instance, the sounds pigs make, such as coughing, can be analyzed for disease detection. A computational system is being developed to monitor and differentiate pig sounds through microphones installed on the farm.
• PCR-Single Strand Conformation Polymorphism was used to determine growth hormone, leptin, calpain, and calpastatin polymorphism in Iranian Balochi male sheep. An artificial neural network model was developed to predict average daily gain in lambs using input parameters of GH, leptin, calpain, calpastatin polymorphism, birth weight, and birth type. The results revealed that the ANN model is an appropriate tool for identifying data patterns to predict lamb growth in terms of ADG given specific gene polymorphism, birth weight, and birth type. The PCR-SSCP approach and ANN-based model analyses may be used in molecular marker-assisted breeding programs to improve the efficacy of sheep production.
• Recent studies by agricultural researchers in Pakistan showed that pro-pesticide state policies have contributed to high pesticide use in cotton crops and reported a negative correlation between pesticide use and crop yield. The excessive use of pesticides is causing a financial, environmental, and social impact on farmers. Data mining within the cotton industry, using pest data along with meteorological recordings, shows how pesticide usage can be optimized.
• A platform of artificial neural network -based models combined with sensitivity analysis and optimization algorithms was used to integrate published data on the responses of broiler chickens to threonine. Analyses of the ANN models for weight gain and feed efficiency suggested that dietary protein concentration was more important than threonine concentration. The results revealed that a diet containing 18.69% protein and 0.73% threonine may lead to optimal weight gain, while optimal feed efficiency may be achieved with a diet containing 18.71% protein and 0.75% threonine.

  Business

• In today's world raw data is being collected by companies at an exploding rate. For example, Walmart processes over 20 million point-of-sale transactions every day. This information is stored in a centralized database, but would be useless without some type of data mining software to analyze it. If Walmart analyzed their point-of-sale data with data mining techniques they would be able to determine sales trends, develop marketing campaigns, and more accurately predict customer loyalty.
• Categorization of the items available in the e-commerce site is a fundamental problem. A correct item categorization system is essential for user experience as it helps determine the items relevant to him for search and browsing. Item categorization can be formulated as a supervised classification problem in data mining where the categories are the target classes and the features are the words composing some textual description of the items. One of the approaches is to find groups initially which are similar and place them together in a latent group. Now given a new item, first classify into a latent group which is called coarse level classification. Then, do a second round of classification to find the category to which the item belongs to.
• Every time a credit card or a store loyalty card is being used, or a warranty card is being filled, data is being collected about the user's behavior. Many people find the amount of information stored about us from companies, such as Google, Facebook, and Amazon, disturbing and are concerned about privacy. Although there is the potential for our personal data to be used in harmful, or unwanted, ways it is also being used to make our lives better. For example, Ford and Audi hope to one day collect information about customer driving patterns so they can recommend safer routes and warn drivers about dangerous road conditions.
• Data mining in customer relationship management applications can contribute significantly to the bottom line. Rather than randomly contacting a prospect or customer through a call center or sending mail, a company can concentrate its efforts on prospects that are predicted to have a high likelihood of responding to an offer. More sophisticated methods may be used to optimize resources across campaigns so that one may predict to which channel and to which offer an individual is most likely to respond. Additionally, sophisticated applications could be used to automate mailing. Once the results from data mining are determined, this "sophisticated application" can either automatically send an e-mail or a regular mail. Finally, in cases where many people will take an action without an offer, "uplift modeling" can be used to determine which people have the greatest increase in response if given an offer. Uplift modeling thereby enables marketers to focus mailings and offers on persuadable people, and not to send offers to people who will buy the product without an offer. Data clustering can also be used to automatically discover the segments or groups within a customer data set.
• Businesses employing data mining may see a return on investment, but also they recognize that the number of predictive models can quickly become very large. For example, rather than using one model to predict how many customers will churn, a business may choose to build a separate model for each region and customer type. In situations where a large number of models need to be maintained, some businesses turn to more automated data mining methodologies.
• Data mining can be helpful to human resources departments in identifying the characteristics of their most successful employees. Information obtained – such as universities attended by highly successful employees – can help HR focus recruiting efforts accordingly. Additionally, Strategic Enterprise Management applications help a company translate corporate-level goals, such as profit and margin share targets, into operational decisions, such as production plans and workforce levels.
• Data mining can be helpful to organizations. Organizational Data Mining is defined as leveraging data mining tools and technologies to enhance organizational decision-making process by transforming data into valuable and actionable knowledge in order to gain a strategic and business competitive advantage. Data obtained – such as employee turnover rates – can help organizations focus their retention efforts accordingly. Additionally, organizational performance management data-mining and analytics applications help firms translate company-level goals, such as profit and sales targets, into operational decisions, as workers KPI and required measured effort levels.
• Market basket analysis has been used to identify the purchase patterns of the Alpha Consumer. Analyzing the data collected on this type of user has allowed companies to predict future buying trends and forecast supply demands.
• Data mining is a highly effective tool in the catalog marketing industry. Catalogers have a rich database of history of their customer transactions for millions of customers dating back a number of years. Data mining tools can identify patterns among customers and help identify the most likely customers to respond to upcoming mailing campaigns.
• Data mining for business applications can be integrated into a complex modeling and decision-making process. LIONsolver uses Reactive business intelligence to advocate a "holistic" approach that integrates data mining, modeling, and interactive visualization into an end-to-end discovery and continuous innovation process powered by human and automated learning.
• In the area of decision making, the RBI approach has been used to mine knowledge that is progressively acquired from the decision maker, and then self-tune the decision method accordingly. The relation between the quality of a data mining system and the amount of investment that the decision maker is willing to make was formalized by providing an economic perspective on the value of "extracted knowledge" in terms of its payoff to the organization This decision-theoretic classification framework was applied to a real-world semiconductor wafer manufacturing line, where decision rules for effectively monitoring and controlling the semiconductor wafer fabrication line were developed.
• An example of data mining related to an integrated-circuit production line is described in the paper "Mining IC Test Data to Optimize VLSI Testing." In this paper, the application of data mining and decision analysis to the problem of die-level functional testing is described. Experiments mentioned demonstrate the ability to apply a system of mining historical die-test data to create a probabilistic model of patterns of die failure. These patterns are then utilized to decide, in real time, which die to test next and when to stop testing. This system has been shown, based on experiments with historical test data, to have the potential to improve profits on mature IC products. Other examples of the application of data mining methodologies in semiconductor manufacturing environments suggest that data mining methodologies may be particularly useful when data is scarce, and the various physical and chemical parameters that affect the process exhibit highly complex interactions. Another implication is that on-line monitoring of the semiconductor manufacturing process using data mining may be highly effective.