Predictive analytics
Predictive analytics uses historical data to predict future events in areas such as finance, meteorology, security, economics, insurance, mobility, and marketing, among others. Generally, historical data is used to create a mathematical model that captures important trends. This predictive model is then applied to current data to predict what will happen next or to suggest actions that will help achieve the best results. Predictive analytics has received a lot of attention in recent years as there have been major advances in assistive technologies, particularly in the areas of big data and machine learning .
definition
Predictive analytics is a sub-discipline and one of the foundations of business analytics in the area of data mining , which deals with predicting future developments. This method has become indispensable, especially with regard to big data, because it offers a tried and tested technique for analyzing large amounts of data and drawing appropriate conclusions. Data mining in particular plays an important role here, because the information received is usually unstructured and must be examined for its usability. Here, a probability for the future is to be calculated, whereby predictive analytics is also used to determine trends. The predictors used - this is a variable in an equation that is used to predict future behavior - can make fairly accurate predictions about the future. The use of multiple predictors then creates a predictive model from which likely events can be calculated.
Process of predictive analytics
- Define Project The project results are defined, including the results, the scope of the effort, the business objectives and the data sets to be used.
- Data collection Data mining for predictive analytics prepares data from several sources for analysis. This provides a complete overview of customer interactions.
- Data analysis Data analysis is the process of reviewing, cleansing, and modeling data in order to discover useful information and reach a conclusion.
- Statistics The statistical analysis enables the assumptions and hypotheses to be validated and tested using standard statistical models.
- Modeling Predictive modeling provides the ability to automatically create accurate predictive models for the future. With the multimodal assessment, the optimal solution can also be selected.
- Deployment Predictive model deployment provides the option to incorporate the analysis results into the day-to-day decision-making process to obtain results, reports and outputs by automating the decisions based on the modeling.
- Model Monitoring Models are managed and monitored to verify model performance and ensure that the expected results are being achieved.
variants
Predictive models
Predictive models use methods from mathematics and computer science to predict an event or outcome. These models predict an outcome in a future state or point in time based on changes in the model inputs. Customers develop the model in an iterative process with a set of training data, then test and validate it to determine the accuracy of its predictions. Different machine learning approaches can be tried to find the most effective model.
The available sample units with known attributes and known performances are referred to as "training samples". The units in other samples with known attributes but unknown performance are referred to as "from the [training] sample". The out-of-sample units are not necessarily chronologically related to the training sample units. For example, the training sample can consist of literary attributes of fonts by Victorian authors with known attribution, and the out-of-sample unit can be a newly found font of unknown authorship; a predictive model can help assign a work to a known author. Another example is the analysis of blood splatters at simulated crime scenes where the entity not included in the sample is the actual blood splatter pattern from a crime scene. The session that is not included in the sample can be from the same time as the training sessions, from an earlier time, or from a future time.
Descriptive models
Descriptive models quantify relationships in data in a way that is often used to classify customers or prospects into groups. Unlike predictive models, which focus on predicting individual customer behavior (e.g. credit risk), descriptive models identify many different relationships between customers or products. Descriptive models do not assign customers based on their likelihood of performing a specific action. Instead, descriptive models can be used to categorize customers, for example, according to their product preferences and their stage of life.
Decision models
Decision models describe the relationship between all elements of a decision - the known data (including the results of predictive models), the decision, and the forecast results of the decision - to predict the results of decisions with many variables. These models can be used for optimization to maximize certain results and minimize others. Decision models are commonly used to develop decision logic or a set of business rules that generate the desired action for each customer or circumstance. Because of the great complexity of real decision problems, there is generally a need to simplify the model. One possibility for simplification is not to take into account all of the characteristics that are considered possible for the decision-relevant data in the model.
Applications
Analytical customer relationship management ( CRM ) is a common commercial application of predictive analysis. Predictive analysis methods are applied to customer data to pursue CRM goals, which are about constructing a holistic view of the customer, regardless of where their information is in the company or in the department concerned. CRM uses predictive analytics in marketing campaign, sales, and customer service applications, to name a few. These tools are necessary for a company to effectively target and focus its efforts on the breadth of its customer base. You need to analyze and understand the products that are in demand or have the potential for high demand, predict customer buying habits to promote relevant products at multiple touchpoints, and proactively identify and mitigate issues that have the potential to attract customers lose or decrease their ability to acquire new customers. Analytical customer relationship management can be applied throughout the customer's lifecycle (acquisition, relationship growth, customer loyalty and recovery).
literature
- Siegel, Eric (2013). Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die (1st ed.). Wiley. ISBN978-1-1183-5685-2.
- Finlay, Steven (2014). Predictive Analytics, Data Mining and Big Data. Myths, Misconceptions and Methods (1st ed.). Basingstoke: Palgrave Macmillan. p. 237. ISBN1137379278.
Individual evidence
- ↑ Stefan Schmitt: Artificial Intelligence: The Vocabulary of the Winner . In: The time . March 17, 2016, ISSN 0044-2070 ( zeit.de [accessed October 18, 2019]).
- ↑ Predictive Analytics: Three Things You Should Know. Retrieved October 8, 2019 .
- ↑ Dr Markus Siepermann: Definition: Predictive Analytics. Retrieved October 8, 2019 .
- ↑ Big data trends at a glance: what is what in predictive analytics? Retrieved October 18, 2019 .
- ↑ What is Predictive Analytics? - Definition from WhatIs.com. Retrieved October 18, 2019 .
- ^ Predictive Analytics. Retrieved October 8, 2019 (German).
- ↑ Predictive Analytics: Three Things You Should Know. Retrieved October 8, 2019 .
- ^ Laux, Helmut .: Decision theory . 4., rework. and exp. Springer, Berlin 1998, ISBN 3-540-64094-0 , pp. 61 ( springer.com [accessed October 9, 2019]).