There is skepticism around the potential of artificial intelligence (AI), which is based on algorithms of machine learning (ML), because of its inability to derive new insights from historical data, make their decisions explainable and transparent, and incorporate new conditions and regulatory frameworks quickly. A way out of this conundrum could be Semantic AI.
At the core of the problem, is the time data scientists spend on collecting and processing uncontrolled data so that data can be explored for useful nuggets. Significantly, many of these efforts focus on building flat files with unrelated data, which lose their relationships with the real world once the features are generated. Therefore, what is needed is a fundamental re-engineering of the underlying architecture, which includes Knowledge Graphs as a prerequisite to calculate not only rules but also corresponding explanations.
Semantic AI, which fuses symbolic and statistical AI, offers the advantages of AI strategies, mainly semantic reasoning and neural networks. Furthermore, Semantic AI is an extension of what is currently used to build AI-based systems. This brings not only strategic options but also an immediate advantage—faster learning from less training data. Semantic AI introduces a fundamentally different methodology and thus additional stakeholders such as knowledge scientists, with complementary skills.
How does Semantic AI methodology make a difference? It offers an alternative approach to develop tools for analysts to access an enterprise Knowledge Graph directly to extract a subset of data that can be quickly transformed into structures for analysis. The results of the analyses themselves can then be reused to enrich the Knowledge Graph.
The Semantic AI approach thus creates a continuous cycle, of which both ML and Knowledge Scientists are an integral part. Knowledge Graphs serve as an interface in between, providing high quality linked and normalized data.
Does this approach offer better results? Apart from its potential to generate broadly accepted explainable AI based on Knowledge Graphs in a trustworthy manner, the use of Knowledge Graphs together with semantically enriched and linked data to train ML algorithms has many other advantages.
The Semantic AI approach leads to results with sufficient accuracy even with sparse training data. It is especially helpful in the cold start phase when the algorithm cannot yet draw inferences from the data because it has not yet gathered enough information.
The approach also leads to better reusability of training data sets, which helps to save costs during data preparation. In addition, it complements existing training data with the background knowledge that can quickly lead to richer training data through automated reasoning. Finally, the approach can help avoid the extraction of fundamentally wrong rules in a particular domain.
Click here to read the original article published here.
Please give your feedback on this article or share a similar story for publishing by clicking here.
