Table 2 Most common algorithms and their business application.
From: Integrating machine learning into business and management in the age of artificial intelligence
Business application | Cluster | Thematic involved | Most used algorithms and technologies |
|---|---|---|---|
Finance | C1—Assessing and forecasting the performance of financial markets | Price clustering Volatility Return predictability Cryptocurrency and currency markets (FOREX) Portfolio optimization and management Investments in emerging economies | Linear discriminant analysis Latent class analysis Causal machine learning Agent-based modelling |
C2—Forecasting macro- and micro-economic indicators | Stock market forecasting Demand and sales forecasting Asset price forecasting | Genetic algorithms Particle swarm optimization Long short-term memory Recurrent neural networks Generative adversarial networks | |
C13—Risk assessment and management | Credit scoring Credit risk assessment Fraud Stress testing Default risk | Automated machine learning Chi-square automatic interaction detection (CHAID) | |
Customer relationship management | C3—Complex network(ing) Analysis | Social networking E-commerce recommendation systems Personalization Document clustering | Graph neural networks Autoencoders Density-based clustering algorithm (DBSCAN) Spectral clustering Affinity Affinity propagation clustering |
C4—Marketing and sales optimization | Customer satisfaction and loyalty Customer retention and churn Benchmarking Segmentation, targeting and positioning Recency, frequency, monetary value (RFM) models and analysis | K-means clustering Principal component analysis Fuzzy [C-means] clustering | |
C8—Opinion mining and eWOM analysis | Customer reviews User-generated content Social media Sentiment analysis Information retrieval | Natural language processing Topic modelling Latent Dirichlet allocation Computer vision | |
Decision-making support | C6—Strategic decision-making | Classification Prediction Regression analysis Decision theory | Random forests Support vector machines Gradient boosting Naïve Bayes classifiers K-nearest neighbours |
C-11—Logistics management | Demand forecasting Revenue management Intelligent transportation systems Game theory Markov decision processes | Spatio-temporal clustering Multi-agent reinforcement learning Debiased machine learning | |
Innovation | C7—Education and skills development | Learning analytics Academic performance Online learning and e-Learning Distance learning Critical thinking Gamification | Latent class clustering Robust clustering Minimum spanning trees Educational data mining |
C9—Innovation and entrepreneurship | Decision-making Competitiveness Entrepreneurship Knowledge spillover | ||
C14—Data management and knowledge discovery | Knowledge management Market basket analysis Knowledge discovery in databases | Intelligent systems Latent semantic indexing Web mining Cross-industry standard process for data mining (CRISP-DM) | |
Public Policy Support | C10—Explainable AI and public policy support | Explainability Accountability Profiling Biases Trust Fairness | Explainable AI Deep neutral networks Multinomial logistic regression |
Other topics and applications | C5—Digital transformation and Sustainability | Business intelligence Predictive analytics Decision support Business analytics Industry 4.0 | MapReduce AdaBoost Distributed clustering Distributed data mining |
C12—GIS applications in Business and Management | Location choice Housing prices Mass appraisal Travel behaviour & mode choice | Spatial clustering Spatial data mining Visual analytics | |
C15—Challenges in data analysis brought by COVID-19 | Ontologies Data integration | Time-series clustering Semantic web Knowledge graphs |
