Human-Computer Interaction and Knowledge Discovery (HCI-KDD): What Is the Benefit of Bringing Those Two Fields to Work Together? [chapter]

Andreas Holzinger
<span title="">2013</span> <i title="Springer Berlin Heidelberg"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/2w3awgokqne6te4nvlofavy5a4" style="color: black;">Lecture Notes in Computer Science</a> </i> &nbsp;
A major challenge in our networked world is the increasing amount of data, which require efficient and user-friendly solutions. A timely example is the biomedical domain: the trend towards personalized medicine has resulted in a sheer mass of the generated (-omics) data. In the life sciences domain, most data models are characterized by complexity, which makes manual analysis very time-consuming and frequently practically impossible. Computational methods may help; however, we must acknowledge
more &raquo; ... hat the problem-solving knowledge is located in the human mind and -not in machines. A strategic aim to find solutions for data intensive problems could lay in the combination of two areas, which bring ideal pre-conditions: Human-Computer Interaction (HCI) and Knowledge Discovery (KDD). HCI deals with questions of human perception, cognition, intelligence, decision-making and interactive techniques of visualization, so it centers mainly on supervised methods. KDD deals mainly with questions of machine intelligence and data mining, in particular with the development of scalable algorithms for finding previously unknown relationships in data, thus centers on automatic computational methods. A proverb attributed perhaps incorrectly to Albert Einstein illustrates this perfectly: "Computers are incredibly fast, accurate, but stupid. Humans are incredibly slow, inaccurate, but brilliant. Together they may be powerful beyond imagination". Consequently, a novel approach is to combine HCI & KDD in order to enhance human intelligence by computational intelligence. Challenges in the Data Intensive Sciences Through the continuing exponential growth of data size and complexity, along with increasing computational power and available computing technologies, the data 320 A. Holzinger intensive sciences gain increasing importance [1]. E-Science is being advanced as a new science along side theoretical science, experimental science, and computational science, as a fundamental research paradigm [2]. Meanwhile, it is established as the fourth paradigm in the investigation of nature, after theory, empiricism, and computation [3], [4]. One of the grand challenges in our networked 21 st century is the large, complex, and often weakly structured [5], [6], or even unstructured data [7] . This increasingly large amount of data requires new, efficient and user-friendly solutions for handling the data. With the growing expectations of end-users, traditional approaches for data interpretation often cannot keep pace with demand, so there is the risk of delivering unsatisfactory results. Consequently, to cope with this rising flood of data, new computational and user-centered approaches are vital. Let us look, for example, at the life sciences: biomedical data models are characterized by significant complexity [8], [9], making manual analysis by the end users often impossible [10] . At the same time, experts are able to solve complicated problems almost intuitively [11] , often enabling medical doctors to make diagnoses with high precision, without being able to describe the exact rules or processes used during their diagnosis, analysis and problem solving [12] . Human thinking is basically a matter of the "plasticity" of the elements of the nervous system, whilst our digital computers (Von-Neuman machines) do not have such "plastic" elements [13] and according to Peter Naur for understanding human thinking we need a different, nondigital approach, one example given by his Synapse-State theory [14] . Interestingly, many powerful computational tools advancing in recent years have been developed by separate communities with different philosophies: Data mining and machine learning researchers tend to believe in the power of their statistical methods to identify relevant patterns -mostly automatic, without human intervention, however, the dangers of modeling artifacts grow when end user comprehension and control are diminished [15] , [16] , [17] , [18] . Additionally, mobile, ubiquitous computing and sensors everywhere, together with low cost storage, will accelerate this avalanche of data [19] , and there will be a danger of drowning in data but starving for knowledge, as Herbert Simon pointed it out 40 years ago: "A wealth of information creates a poverty of attention and a need to allocate that attention efficiently among the over abundance of information sources that might consume it" [20] . Consequently, it is a grand challenge to work towards enabling effective human control over powerful machine intelligence by the integration of machine learning methods and visual analytics to support human insight and decision support [21], the latter is still the core discipline in biomedical informatics [8] . A synergistic combination of methodologies, methods and approaches of two areas offer ideal conditions for addressing these challenges: HCI, with its emphasis on human intelligence, and KDD, dealing with computational intelligence -with the goal of supporting human intelligence with machine intelligence -to discover new, previously unknown insights within the flood of data. The main contribution of HCI-KDD is, following the notion: "science is to test ideas, engineering is to put these ideas into business" [22] , to enable end users to find and recognize previously unknown and potentially useful and usable information. It may be defined as the process of identifying novel, valid and potentially useful data patterns, with the goal of understanding these data patterns for decision support.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-642-40511-2_22">doi:10.1007/978-3-642-40511-2_22</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/nluwieg3sjcxteiotjom6jskre">fatcat:nluwieg3sjcxteiotjom6jskre</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190503072454/https://link.springer.com/content/pdf/10.1007%2F978-3-642-40511-2_22.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/80/7e/807e779f04606cabadd753b5827bfe91d7b83a6c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-642-40511-2_22"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a>