Aktualisiert: Okt 23
Content: A close look to the differences between the terminologies Business Intelligence, Business Analytics, Advanced Analytics, Machine Learning, Deep Learning and Artificial Intelligence.
Introduction: With all the developments and technical advances, it is not easy not to lose the thread. The more important it is to think about the following terms in order to be sure to use them correctly in everyday business.
In the following, we will look in detail at what the individual terms make up so that an understanding can be formed as to which field is involved when:
What is business intelligence?
Accordning to the Amercian market research company Forrester Research, the definition of Business Intelligence (BI) is:
"Business intelligence is a set of methodologies, processes, architectures, and technologies that transform raw data into meaningful and useful information used to enable more effective strategic, tactical, and operational insights and decision-making."
So one could say that it can be defined as the process of data gathering, data storage and knowledge management in general. This contains topics like the integration of the data, ensuring the datas quality, data warehousing, master-data management, text-analytics and last but not least content-analytics. Some specific elements of business intelligence would be:
- The optimization of key performance indicators
- Multidimensional aggregation and allocation
- Realtime reporting including an analytical alert
- Tagging, standardization and denormalization
- Group consolidation, budgeting, and rolling forecasts
- Statistical inference and probabilistic simulation
- Its considered a method of interfacing with unstructured data sources
- Process management and version control
- Open item management
Summarized business intelligence is: The process of knowledge management, where data is gathered and stored. This raw data is transformed into meaningful information, trough analytics. And the process ends with reporting and visualization with e.g. dashboards, with the ultimate goal to support in the decision making process.
What is Business Analytics?
Here we use past business data and try to get insights, to ensure future business success. The process of gathering a better business understanding is pursued through statistical methods. Numerical analysis, as well as analytical modelling, explanatory modelling and predictive modelling, help in the decision making process.
- Behavioral analytics
- Cohort analysis
- Competitor analysis
- Cyber analytics
- Enterprise optimization
- Financial services analytics
- Fraud analytics
- Health care analytics
- Key Performance Indicators (KPI's)
- Marketing analytics
- Pricing analytics
- Retail sales analytics
- Risk & Credit analytics
- Supply chain analytics
- Talent analytics
- Transportation analytics
- Customer Journey Analytics
- Market Basket Analysis
Summarized business analytics is: The use of past business data to get, through the method of statistical evaluation, new business insights, which help in the decision making process.
What is the difference between business intelligence and business analytics?
If questions are answered of what happened, how often, how many, where the problem is and what actions are needed, then we are talking about reporting, queriying, online analytical processing, or just alert tools, which means we are talking from the area of business intelligence.
When questions are asked of why something is happening, what if such trends will continue, what do one has to assume will happen next and what would be the best outcome that could happen, we are talking about predictions and optimizations, which means we are talking about the area of business analytics.
What are advanced analytics?
If deeper insights, recommendations or predications are wanted and an autonomous or semi-autonomous examination of the data is possible, we talk about advanced analytics. This means that technology beyond the above explained business intelligence is used, like:
- data or text mining
- pattern matching
- machine learning
- semantic analysis
- network and cluster analysis
- sentiment analysis
- multivariate statistics
- graph analysis
- complex event processing
- neuronal networks
What is machine learning?
Machine learning is a subset from artificial intelligence and is seen as the study of computer algorithms, which have the ability to improve automatically by themselves through experience. Based on sample data, which is also known as training data, the machine learning algorithm builds a mathematical model to be able to make predictions or decisions. The algorithm will not be explicitly programmed to do so, but it will be able after sufficient training.
What is Deep Learning?
Deep learning is part of the machine learning family. It is based on neuronal networks and there are different methods: supervised, unsupervised or semi-supervised learning. There are different deep learning architectures:
- Deep neural networks
- Deep belief networks
- Recurrent neural networks
- Convolutional neural networks
The networks are applied in different fields, for example:
- machine vision (image search, image edition and enhancement: machine learning, facial recognition applications: authentication system, data efficient home security: IoT, interacting with the real world: augmented reality, medical image processing: cancer detection and prediction, radiology and diabetic retinopathy, playing games: reinforcement learning, cashier-less stores: Amazons store Go, Self-driving cars, surveillance, autonomous weapons: computer visions is able to give eyes to weapons by identifying objects and chosing targets)
- computer vision (image classification, image classification with localization, object detection, object segmentation, image style transfer, image colorization, image reconstruction, image super-resolution, image synthesis)
- speech recognition (recognition and translation of spoken language into text by computers)
- natural language processing (Information extraction, relation extraction, named entity recognition, part of speech tagging, coreference resolution, Parsing, Word Sense Disambiguation)
- audio recognition (audio signal recovery, speech quality enhancement, nonlinear transducer linearization, learning based pseudo-physical sound synthesis)
- machine translation (Google translate, Microsoft translator, DeepL)
- social network analysis (prediction of the performance of recommender systems)
- drug design (analysis of structure–activity data, establishment of quantitative structure–activity relationships (QSAR), gene prediction, locating protein-coding regions in DNA sequences, 3D structure alignment, pharmacophore perception, docking of ligands to receptors, automated generation of small organic compounds, and the design of combinatorial libraries)
- bioinformatics (protein structure and function prediction)
- medical image analysis (example cancer)
- material inspection (example textile industry)
- board game programs ( example AlphaGo)
The use of multiple layers within the neuronal network gives the name "deep learning" its name, as the word "deep" is referring to the different layers.
In the next blogpost we will dive deeper and we will have a quick look at machine learning and its different components: Unsupervised and supervised learning, problem classes, algorithms and evaluation methods.