Analytics and Advanced Analytics
*OLAP is a data processing technology that consists in the preparation of aggregate (aggregated) information based on large data sets structured on a multidimensional principle. OLAP implementations are components of business intelligence (BI) class software solutions.
*BPM (Business Process Management, Business Process Management) is an organization management concept that represents the activities of an enterprise as a set of processes. combines the ideas of business process management and the IT environment for their change (specialized software, BPM-system) using BPMN notation.
Business intelligence — BI:BI is the use of computational technologies to identify, detect and analyze business data such as sales revenue, products, costs, revenues, etc. BI technologies provide current, historical and predictive representations of internally structured data for products and units, enabling more efficient decision-making and strategic operational information through functions such as operational analytical processing (OLAP), reporting, predictive analytics, data/text mining, comparative analysis, and business performance management (BPM).
Descriptive analytics:As an analyst or business owner, when you’re looking for an answer to a question, What’s going on in my business? That’s where descriptive analytics comes in. It’s the most common and widely used analytics technique that analyzes real-time data, usually using effective visualization tools like dashboards, and allows us to learn from past behavior and give insight into how it will affect future results. But because it only gives us an idea of whether everything is good in our business, but doesn’t explain the root cause of it. For this reason, data-driven businesses combine descriptive analytics with other types of data analytics to find a comprehensive solution.
Diagnostic analytics:When you already know what’s going on in your business using descriptive analytics, you want to know the answer to the next question. Why is this happening in your business? If you want to know the root cause of this, it’s where the diagnostic analyst plays a role, helping analysts or data scientists delve into the data to find the answer, and typically in business, BI dashboards help to detail information through hierarchies or make quick comparisons to find causes or factors that affect the business.
Predictive analytics:Predictive analytics is based on the results of descriptive and diagnostic analytics and is used to search for answers to the question: What is likely to happen in the future based on previous trends and patternsIn general, it’s all about prediction. Predictive analytics uses various statistical algorithms and machine learning algorithms to provide recommendations and answers to questions related to what might happen in the future that BI can’t answer. But because it’s probabilistic, it just gives an estimate of a possible future outcome. Also, the accuracy isn’t 100%, because it all depends on the quality of the data, how you make informed assumptions about missed values, and how optimization is performed.
Prescriptive analytics:When you get the results of descriptive, diagnostic, and predictive analytics, such as what happened, the root cause of it, and what might happen in the future, the Prescription Model uses those answers to help you. Identify the best way to bypass or eliminate future problemsYou can use prescriptive analytics to advise users about possible outcomes and what they should do to maximize key business metrics. The best example is Yandex Maps, which will help you choose the optimal route based on distance, traffic and speed.
Machine Learning / Deep Learning (ML):Machine learning is broadly interpreted as giving computer systems the ability to “learn.” ML’s goal is to enable machines to learn on their own using the data provided and make accurate predictions. ML is a subset of artificial intelligence; in fact, it’s just a method of implementing AI. It’s a method of training algorithms so they can learn how to make decisions. Machine learning involves transferring large amounts of data to an algorithm and allowing it to learn more about the information processed.
Deep Learning (DL) is a subset (ML): in fact, it’s just a method of implementing machine learning. In other words, DL is the next evolution of machine learning. DL algorithms are largely based on information processing patterns found in the human brain. Just as we use our brains to identify patterns and classify different types of information, deep learning algorithms can be taught to perform the same tasks for machines. Normally, the brain tries to decipher the information it receives. This is achieved by labeling and categorizing elements. Whenever we receive new information, the brain tries to compare it to a known element before making sense of it — the same concept that deep learning algorithms use.
Artificial Intelligence / Cognitive Analytics:Cognitive analytics combines a number of intelligent technologies, such as artificial intelligence, machine learning algorithms, deep learning, similar to the human brain, to perform certain tasks..Essentially, this type of analytics is based on how the human brain processes information, draws conclusions, and codifies instincts and experiences in learning, such as understanding not only the words in a text, but the entire context of what is written or spoken. All of these smart technologies over time make a cognitive application smarter and more efficient by learning from its interactions with data and people.
| Attributes | Business intelligence | Descriptive | Diagnostics | predictive | Prescriptive | Machine / Deep Learning | Artificial intelligence / Cognitive |
| Perspective | Past | Past | Past | The future | The future | The future | The future |
| Type of questions | What happened? | What happened? | Why did this happen? | What could happen | What should we do? | The Reason Something Should Happen | How to Improve Human Thinking |
| Method and techniques | Reporting, monitoring, alerts, dashboard, application maps, OLAP and Adhoc queries | Data mining, data discovery, OLAP, Adhoc queries, monitoring panels | Semantic analysis and tonality, data mining, modeling statistics, OLAP, Decision Tree | Predictive modeling, Neural networks, Comparison with sample, Forecasting, Regression analysis, Simulation, Alerts | Optimization Models, Heuristics, Discrete Choice Modeling, Linear/Nonlinear Programming, Value Analysis, Schedule Analysis | Natural language processing, machine and deep learning, training data, evaluation, labeling, anomaly detection | Cognitive Consultants, Artificial Intelligence/DL, Automatic Resolution |
| Data data | Structured | Structured | Structured and unstructured | Structured, unstructured and semi-structured | Structured, unstructured and semi-structured | Structured, unstructured and semi-structured | Structured, unstructured and semi-structured |
| Knowledge generation | Management | Management | Management | Automatic. | Automatic. | Automatic. | Automatic. |
| Users | Business users | Business users | Business users | Data Processor, Business Analyst, Business Users | Data Processor, Business Analyst, Business Users | Data Processor, Business Analyst, Business Users | Data Processor, Business Analyst, Business Users |
| Business initiatives | Jetty | Jetty | Jetty | Proactive | Proactive | Proactive | Proactive |
| Results | Table | Table | Table | Table | Table | Answer. | Answer. |
| Area of application | Unlimitedly. | Unlimitedly. | Unlimitedly. | Unlimitedly. | Unlimitedly. | A specific business issue | A specific business issue |
| Examples of Platforms/Tools | SAP BI, Cognos BI, Microstrategy BI, SAS BI, QlikView, Tableau BI, JasperSoft BI | SAP BI, Cognos BI, Microstrategy BI, SAS BI, QlikView BI, Tableau BI, JasperSoft BI | SAP, Cognos, SAS, R Enterprise, Tableau, JasperSoft | RapidMiner, KNIME, SAP Predictive Analytics, IBM Predictive Analytics, Microsoft R, SAS Predictive Analytics | SAP HANA, IBM SPSS, RapidMiner Studio, SAS Advanced Analytics, Radius | Google ML, Amazon ML, Amazon SageMaker, Accord.NET, Azure ML | Microsoft Cognitive Toolkit, Keras, TensorFlow, Theano, Caffe |
| Programming language | R, Python, Java, SQL | R, Python, Java, SQL | R, Python, Java, SQL | R, Python, Java | R, Python, Java | Pyton, C / C ++, Java, R, Javascript, TensorFlow | Python, R, Prolog, JAVA, C++ and LISP |
| Examples of use | Reporting, dashboard | Measurement, monitoring, KPI, | Trend Analytics, Situational Analysis, First Cause, Cluster Analysis, Navigation, OLAP, Flexible and Spatial Visualization | Forecasting, probability assessment, recommendations for risk management | Scenario Planning, Strategy Formulation and Optimization, Recommendations, Rule Systems, BPM Automation | Transformation of speech into text, Transformation of text into speech, NLP, Explanation, Translation, Sentiment Analysis | Image/video classification, Facial recognition, Authentication, Visual recognition, Process control, Gesture management, Robotics, Common Sense |
| Experience | IT and business users | Information Technology and Business Users, Business Analyst | Business users, Business analyst | Business Users, Business Analyst, Data Processor | Business Users, Business Analyst, Data Processor | Business Users, Business Analyst, Data Processor | Business Users, Business Analyst, Data Processor |
| Complexity of integration | Moderate. | Moderate. | Moderate. | Tall. | Tall. | Tall. | Tall. |
| Architecture | Package | Package | Package | Package management / In real time | Package management / In real time | In real time. | In real time. |
| Restrictions | Accumulation of historical data, cost, complexity, limited use, time-consuming implementation | Accumulation of historical data, cost, complexity, limited use, time-consuming implementation | Accumulation of historical data, cost, complexity, limited use, time-consuming implementation | Expertise, implementation, end-user empowerment, burdensome lists of projects | Expertise, implementation, end-user empowerment, burdensome lists of projects | Provability, data privacy and security, algorithm bias, data scarcity | Provability, data privacy and security, algorithm bias, data scarcity |