The Effect of Hyperparameter Selection on the Personification of Customer Population Data

We explore the effects of hyperparameter selections on the personification accuracy of customer analytics data from a corporate YouTube channel with an audience in the hundreds of thousands and customer interactions in the tens of millions. Using non-negative matrix factorization, we generate personas sets from 5 to 15 using the customer analytics data, with the number of personas being the changing hyperparameter. We then compare the gender, age, nationality, and topical interests of the personas across each of the 11 persona sets using the average of the 110 generated personas as the baseline. This analysis shows that hyperparameter selection significantly alters the personification of the analytics data, with the effect most apparent with age representation. The set of 10 personas provides one of the most accurate representations across all attributes, indicating that this may be a good default hyperparameter for personification. Future research can explore other personification attributes with other customer analytics datasets.

A Systematic Review of the Literature on Customer Analytics to Predict Dynamic Purchase behavior of Consumers through Credit Cards

The Credit card industry is flooded with customer data and also their diversified purchasing behavior. These multifariousness analysis helps different industry to maximize revenue through data and to identify right cluster of audience based on the products and services they offer. To analyse this data efficiently and erect efficient co-relation and streamline process execution, expertise in customer segmentation specifically through customer analytics, is required. This domain of analytics uses a wide variety of concepts related to data mining, Passive and active data collection, Media planning, Regression analysis, Customer based corporate-valuation, Market-structure, Probability Models, Optimization, visualization and Implementation of model spreadsheet for predictions on data set. Moreover, customer analytics is the subset of a larger range of business analytics techniques used for better statistical and predictive analysis to transform and make smarter, data-driven business decisions. Keywords: Regression analysis, Market-structure, Optimization, Business analytics, Model spreadsheet.

Customer reviews analytics on food delivery services in social media: a review

Food delivery services have gained attention and become a top priority in developed cities by reducing travel time and waiting time by offering online food delivery options for a variety of dishes from a wide variety of restaurants. Therefore, customer analytics have been considered in business analysis by enabling businesses to collect and analyse customer feedback to make business decisions to be more advanced in the future. This paper aims to study the techniques used in customer analytics for food delivery services and identify the factors of customers’ reviews for food delivery services especially in social media. A total of 53 papers reviewed, several techniques and algorithms on customer analytics for food delivery services in social media are Lexicon, machine learning, natural language processing (NLP), support vector machine (SVM), and text mining. The paper further analyse the challenges and factors that give impacts to the customers’ reviews for food delivery services. These findings would be appropriate for development and enhancement of food delivery services in future works.

Current Directions and Future Research Priorities of Customer Data Analysis

Customer analysis is receiving special attention from both researchers and professionals. The objective of this paper is to identify the trends of techniques used to address customer’s current problems and shed light on future research directions using a literature review. We reviewed the literature for the last five years. The findings revealed that customer purchase was the most popular technique used by the research community followed by customer satisfaction and visit wit. Whereas customer segmentation and customer churn were the least. However, the regression method was commonly used for predicting customer purchase and behavior. But, social media and big data are still in their early stages for customer analytics research.

Data science in the business environment: customer analytics case studies in SMEs

Purpose A vast amount of complex data is being generated in the business environment, which enables support for decision-making through information processing and insight generation. The purpose of this study is to propose a process model for data-driven decision-making which provides an overarching methodology covering key stages of the business analytics life cycle. The model is then applied in two small enterprises using real customer/donor data to assist the strategic management of sales and fundraising. Design/methodology/approach Data science is a multi-disciplinary subject that aims to discover knowledge and insight from data while providing a bridge to data-driven decision-making across businesses. This paper starts with a review of established frameworks for data science and analytics before linking with process modelling and data-driven decision-making. A consolidated methodology is then described covering the key stages of exploring data, discovering insights and making decisions. Findings Representative case studies from a small manufacturing organisation and an independent hospice charity have been used to illustrate the application of the process model. Visual analytics have informed customer sales strategy and donor fundraising strategy through recommendations to the respective senior management teams. Research limitations/implications The scope of this research has focused on customer analytics in small to medium-sized enterprise through two case studies. While the aims of these organisations are rather specific, they share a commonality of purpose for their strategic development, which is addressed by this paper. Originality/value Data science is shown to be applicable in the business environment through the proposed process model, synthesising micro- and macro-solution methodologies and allowing organisations to follow a structured procedure. Two real-world case studies have been used to highlight the value of the data-driven model in management decision-making.