Discussion

Response 1:

Visual Analytics is a process to transform the approach information into an opportunity in a way as information visualization has changed our view on databases, the purpose of Visual Analytics is to make the frameworks of taking care of data and information clear for an analytic discourse. Visual Analytics helps in propelling the gainful evaluation, change and quick improvement of the strategies and models improving the data for choosing better decisions. The degree of visual Analytics can in like manner be depicted similar to the merged information and communication technology (ICT) key advances like information visualization, data mining, knowledge discovery or illustrating, and reenactment" (Janssen, Wimmer and Deljoo, 2015, p. 325). The two important parts of the model given by Keim,2008 are information visualization (upper part) and automated data analysis (lower part). The four pieces of visual data examination are visualization, Data, Models and knowledge.

Here I would like to discuss the process between visualization and models, in perspective on data volume and complexity information visualization can't be honestly applied. Here comes the necessity for data analytics. Visualization visualizes the changed data in the structure where the customer can get data and assist in building models. At the point when the model is made the parameters which are ought to have been modified are then changed and is given to the data which is then moved to the Data where the additional data mining is finished and the yield of data mining is given to the models. Right, when all-out data mining is finished the results are given from the models to portrayal for model discernment. This system of coordination of visual and customized data assessment technique is valuable for expansive and keen decision help.

Visual analytics actions help development that joins the characteristics of human and electronic data planning, discernment transforms into the techniques for a semi-robotized informative strategy, where individuals and machines take an interest using their specific undeniable capacities for the best results (Keim, 2008).

Response 2:

Visual data exploration may appear Analytics 101, yet experts who avoid this progression may pass up significant bits of knowledge and a more profound comprehension of the data with which they are working.

        On the off chance that something glances wrong in your data, it presumably isn't right, said Tatiana Gabor, an investigation supervisor for the income group at music gushing organization Spotify. Visual data disclosure instruments reliably rank among examination buyers' top needs. Be that as it may, the product is regularly conveyed as an end unto itself, with numerous organizations buying it to work as a self-administration investigation apparatus for business clients. In the hands of experienced data researchers, in any case, it can create much more profound bits of knowledge.

     Data exploration is a suggested initial phase in any investigation, yet examiners frequently simply take a gander at numbers: rundown measurements like mean, middle and spread. They don't generally take part in visual data exploration.

        A few examiners additionally carry a lot of suspicions to data and test those immediately by running the data through a relapse or grouping model. Yet, hopping to these strategies initially can make an examiner disregard significant highlights of the data. Data visualization is a basic device in the data investigation process. Visualization errands can extend from creating key appropriation plots to understanding the interaction of complex powerful factors in AI calculations. In this instructional exercise, we center around the utilization of visualization for beginning data investigation.

Data visualization is a basic apparatus in the data examination process. Visualization undertakings can extend from creating crucial appropriation plots to understanding the interaction of complex powerful factors in AI calculations. In this instructional exercise, we center around the utilization of visualization for starting data investigation.

       Visual data investigation is a required initial step whether progressively formal examination pursues. At the point when joined with expressive insights, visualization gives a viable method to recognize synopses, structure, connections, contrasts, and variations from the norm in the data. In many cases, no detailed investigation is fundamental as all the significant determinations required for a choice are apparent from basic visual assessment of the data. Different occasions, data investigation will be utilized to help control the data cleaning, include determination, and examining process.

         In any case, visual data investigation is tied in with researching the attributes of your data set. To do this, we ordinarily make various plots in an intelligent manner. This instructional exercise will tell you the best way to make plots that answer a portion of the basic inquiries we commonly have of our data.

References:

Bowen Yu, Claudio T. Silva (2019). Florence: A Natural Language Interface for Visual Data Exploration within a Dataflow System

   Submitted on 2 Aug 2019 ( v1 ), last revised 6 Oct 2019 (this version, v2)

Zhe Cui, Sriram Karthik Badam, Adil Yalçin, Niklas Elmqvist(2018). DataSite: Proactive Visual Data Exploration with Computation of Insight-based Recommendations

   Submitted on 23 Feb 2018 ( v1 ), last revised 22 Sep 2018 (this version, v3)

Response 3:

Information visualization emerged from research in human-computer interaction, computer science, graphics, visual design, psychology, and business methods(Shneiderman and Bederson 2003). It allows to intuitively access results of complex models, even for nonexperts, while not being limited to intrinsic application fields. In fact, information visualization is increasingly considered as critical components scientific research, data mining, digital libraries, financial data analysis, manufacturing production control, market studies, and drug discovery (Shneiderman andBederson 2003). The growing amount of data collected and produced in modern society contains hidden knowledge that needs to be considered in decision making. Due to the data connecting the information visualization (top) and the data mining (bottom) processes volume and complexity information, visualization can no longer be applied alone. A new research discipline within information visualization was introduced. Visual analytics is defined as “the science of analytical reasoning facilitated by interactive visual interfaces” (Thomas and Cook 2005). The goal of visual analytics research is the creation of tools and techniques to enable the user to (a) synthesize information and derive insight from massive, dynamic, ambiguous, and often conflicting data, (b)detect the expected and discover the unexpected, (c) provide timely, defensible, and understandable assessments, and (d) communicate assessment effectively for action. In contrast to pure information visualization, visual analytics combines interactive visualization with automated data analysis methods to provide scalable interactive decision support. Figure 15.2 shows an adaptation of Keim’s widely accepted process model for visual analytics (Keim et al. 2008). The visual data exploration process from information visualization (upper part), and automated data analysis methods (lower part) are combined with one visual, and interactive analysis process model. The users directly included in the model by interactive access to the process steps. This generic process model makes visual analytics applicable to a variety of data-oriented research fields such as engineering, financial analysis, public safety and security, environment and climate change, as well as socioeconomic applications and policy analysis, respectively. The scope of visual analytics can also be described in terms of the incorporated information and communication technologies (ICT) key technologies like information visualization, data mining, knowledge discovery or modeling, and simulation (Keim et al. 2008). In its framework program seven, the European Commission (EC) emphasized visualization as a key technology in the objective for ICT for governance and policy modeling (European Commission 2010). Recently, methodologies on how to design and implement information visualization and visual analytics solutions for data-driven challenges of domain specialists have been presented (Munzner 2009; Sedlmair et al. 2012). Due to their reflection upon practical experiences of hundreds of information visualization and visual analytics research papers, the value of the introduced methodologies is widely recognized. In these methodologies, visualization researchers are guided in how to analyze a specific real-world problem faced by domain experts, how to design visualization systems that support solving this problem, and how to validate the design. Considering information visualization validation, we refer to Lam et al. (2012). Recent approaches in visual analytics focus on the questions on how to simplify the access to the analysis functionality of visual analytics techniques, and on how to present analysis results. This includes the analysis process with its intermediate steps, and the findings derived from the visual analytics techniques (Kosara and Mackinlay2013).