Artificial intelligence in business - what can it do and what are its limitations? And how can artificial intelligence be easily integrated into common business software?

A common understanding of artificial intelligence (AI) describes the attempt to teach machines intelligent behavior. It encompasses diverse approaches not only from computer science, but also from fields such as engineering, logic and statistics from mathematics, linguistics, and many more [1]. Usually, algorithms form the basic framework for this intelligence. Neural networks are a prominent example of such algorithms in AI.

But what about the danger from a superior AI prophesied in science fiction movies? This thought refers only to a part of AI research: "strong" or also " general" AI describes the development of an artificial intelligence, which would be at least equal to humans [1]. A thought that may seem disturbing, but is far from technically possible [2].

In contrast, the second field of AI is very much within the realm of possibility: However, so-called "weak" AI, as it is readily taken up in business and many research fields, can only perform well-defined tasks. However, the "weak AI" algorithms are already very successful for this purpose:

For example, such AI applications can defeat humans in chess, other AI applications can successfully interpret situations in road traffic in autonomous driving, or still other applications can do translations [2]. It is important to keep in mind that such applications still only derive rules based on very large sets of example situations to answer limited, immediate and practical questions.

From a business perspective, these impressive capabilities strengthen the desire to use AI for commercial purposes as well. However, this is contrasted by the fact that SMEs still often lag behind large companies, especially when it comes to sophisticated digital applications such as in data analysis. [3]. Certainly, it is not the goal of every SME to set up its own data science department, but that does not mean that information and business benefits have to be lost in unused data.

This series of articles will provide insight into how Analytical Process Automation (APA) software can be used to make optimized business decisions based on data.

The APA software "Alteryx" used here offers companies the possibility to develop and automate data processing procedures without prior programming knowledge. In addition, Data Scientists can embed self-developed code as an application and thus collaborate smoothly with business users.

Introduction to Analytic Process Automation: Analytics + Data Science + Process Automation

Outlook:

In article 2 follows: Using Alteryx Software from a Data Scientist's Perspective: using a Python plugin in the "Alteryx" software, a simple neural network for image recognition is trained and made available to other business users.

In article 3 follows: The use of the Alteryx software from the point of view of a business user: If a neural network was made by Data Scientists, the software "Alteryx" enables other business users to use such prefabricated neural networks themselves in an elegant way, without having to delve into Python or R - code.  In addition, an exemplary application illustrates the limitations of AI.

In article 4 follows: So in what areas can the strengths of AI be leveraged in SMEs? A typical implementation process flow is presented. 

Bibliography:

[1] Nilsson, Nils J. (2009). The Quest for Artificial Intelligence. A History of Ideas and Achievements. Cambridge University Press.

[2] Spiegelhalter, David (2019). The art of statistics: learning from data. Penguin UK.

[3] Zukunft der KMU liegt in der Digitalisierung