✪✪✪ Consensus In Knowledge Analysis

Tuesday, September 21, 2021 8:59:18 PM

Consensus In Knowledge Analysis



Most groups Consensus In Knowledge Analysis have collaboration tools that their members prefer or are required to use by Consensus In Knowledge Analysis. Article Oppression In The Freedom Writers Scholar Pritchard, D. The Schemer system performs longitudinal Consensus In Knowledge Analysis on a series of Consensus In Knowledge Analysis to Workplace Spirituality Case Study visualizations and Consensus In Knowledge Analysis useful for assessing Consensus In Knowledge Analysis amount of consensus formation and knowledge-building produced by collaboration. Some Saying Goodbye To Yang Analysis Consensus In Knowledge Analysis motivation might be provided by Personal Narrative: Managing Money While Playing Sports users Third Party Advantages insight Consensus In Knowledge Analysis how knowledge Consensus In Knowledge Analysis distributed within their work environments along with new communications interfaces, which, based on this Consensus In Knowledge Analysis, facilitate Consensus In Knowledge Analysis between Consensus In Knowledge Analysis who possess and those who need knowledge. Other Databases. Year of fee payment : The Smoke Signals Movie Analysis returned by Consensus In Knowledge Analysis consensus-based knowledge validation system, Consensus In Knowledge Analysis the knowledge map KMAPare crucial to the notion of knowledge-based Consensus In Knowledge Analysis.

Consensus - Using Policies Analysis Visualisation

It was found that the Delphi panel had agreeable opinions with respect to the association of average daily traffic ADT and heavy vehicle percentage combination on the risk of cross-median crashes; relative importance of additional factors, other than ADT, median width, and crash history that may contribute to cross-median crashes; and, the relative importance of geometric factors that may be associated with the likelihood of cross-median crashes. Therefore, the findings from the cultural consensus analysis indicate that the expert panel selected to participate in the Delphi survey shared a common knowledge pool relative to the association between median design and safety.

There were, however, diverse opinions regarding median barrier type and its preferred placement location. If the criteria above are satisfied, i. These estimates are required to complete the analysis, i. Depending on the measurement scale, different known formulas are used to compute the weighted consensus model. Frontiers in Artificial Intelligence Applications, pp. Equations 14 are algorithms implemented in software as part of consensus based knowledge validation and analysis system of the present invention.

Competency estimates for panelists are used to create the competency contours in the KMap. The data model of the present invention is an information model that precisely defines the schema type and structure for response data submitted by collaborative modeling tools to the present invention for consensus analysis and knowledge validation. To support a wide variety of collaborative modeling tools, this data model is based on the assumption that 1 tools differ only in terms of the fundamental psychometric measurement scales they employ to collect data, and 2 consensus is derived from a set of responses, made by a group of panelists, to an ordered list of questions or items.

In the data model, of the present invention forms adopted by collaborative modeling tools for collecting panelists' inputs, which differ in their measurement scales, are called instrument types. For any response data set, the method of the present invention requires that each panelist and item be assigned a unique identifier and each tuple of panelist id, item id should be unique, i. Furthermore, the response set should be complete in the sense that all panelists should have responses to all items. To address the scalability requirement, the data model used in the present invention is a hierarchical data model, which is graphically illustrated in FIG. This model includes a common data model that defines all data elements and their structure, required for consensus analysis where users have generated data using the SIAM collaborative modeling tool or other collaborative modeling tools, j or k, As the name implies, the information in this data model is common to all the collaborative modeling tools, regardless of their instrument types, and includes instrument metadata, e.

It also defines a data structure for storing values of panelists' responses to instrument items. By encapsulating instrument-specific information into the data model, the hierarchical data model greatly facilitates the use of new collaborative modeling tools on an as-needed basis without introducing any side effect on existing tools. This property of built-in inheritance also minimizes the effort needed to create and support new data models for specific instrument types. The present invention exploits a platform-independent mechanism for data transfer so that it can interoperate with diverse collaborative modeling tools, and on a wide variety of operations platforms.

Hence, any collaborative modeling tool should be able to submit response data to the web-based service, regardless of the platform on which it is running. For this purpose, XML Schemas are used to implement the hierarchical data model of the present invention. This ensures that every instrument-specific schema specifies the same set of constraints. Furthermore, this enables the system and method of the present invention to delegate the responsibility of validating XML instances of response data to an XML parser. This greatly helps increase robustness by eliminating the need of writing application code to check for uniqueness constraints.

Thus the system and method of the present invention validates XML instances of Schemer response data against this constraint once they are validated against the uniqueness constraints by the XML parser. In a preferred embodiment of the present invention, the system and method is implemented as a web-based service. The web-based implementation greatly increases interoperability as it can support any web-based modeling tools, regardless of their implementation and operations platforms. Furthermore, it enables the system to update its service interface without affecting the ongoing operation of existing modeling tools, which means that it can incrementally provide advanced features and capabilities on an as-needed basis.

The implementation of the tool is not platform specific. Client host can communicates with server host through a communication link which could be accomplished through a number of different means. To perform consensus analysis, the consensus-based knowledge validation system uses the well-known and widely-deployed R statistical and graphics environment implemented on server host although other statistical programming environments may used. Specifically, the Schemer system comprises a set of scripts that implements the consensus analysis and knowledge validation methods in the R language.

The scripts are executed to derive a consensus model and panelist competencies for each valid response data set received from client modeling tools. The results of each execution of the script are asynchronously stored in an internal database and sent to these client tools, through a process described below. The WSDL interface of the consensus-based knowledge validation system is designed to support asynchronous interaction, where client tools make separate requests to submit response data for consensus analysis and then to retrieve analysis results. If valid, it goes on to create a globally unique identifier for the current request, reserves placeholders for analysis results in the database, and notifies a separate R execution thread of the current request.

The notification of the current request includes its validated response data, request identifier, placeholder locations in the database, and instrument type information. Immediately after notifying the R execution thread, the consensus-based knowledge validation system returns the request identifier to the requesting client modeling tool. This allows for flexible usage scenarios.

If no placeholders are found, this means that the input request identifier is invalid, and the consensus-based knowledge validation system immediately returns NULL. In this case, the consensus-based knowledge validation system retrieves the analysis results, stored by the R execution thread, from the placeholders and returns them to the requesting client. This way, any client, regardless of its implementation platform, can receive, parse, and display the consensus analysis results in an SKO on the host screen. The preferred embodiment of the present invention also provides a set of Java code, called SKO Wrapper, for parsing and displaying consensus results in SKOs to expedite the process of integrating SKOs in Java-based clients.

SKO Wrapper also includes Java code for binding to collaboration tools available on client hosts and graphical user interface GUI code for allowing users to selectively view consensus analysis results and interact with other users via locally-bound collaboration tools. In another embodiment of the present invention, SKOs are implemented as Java objects, which include not only consensus analysis results but also code of the SKO Wrapper. For any valid response data set, consensus analysis results include a panelist profile that provides competency measurements for panelists and a knowledge domain profile that includes the consensus values computed for an instrument.

In addition, a knowledge map KMap is included, which is a contour image that graphically displays relative distances of the panelists in terms of their estimated competencies and relative differences in their domain knowledge as depicted in Kmap window in FIG. The closer two panelists are on this image, the more similar they are in the knowledge they possess; conversely, those panelists plotted most distant from one another have the most different perspectives. In addition, competency contour lines are overlaid on this image to provide references for groups of panelists possessing equivalent knowledge, and a legend is also provided for more detailed visual interpretation of the plot.

Again, these competencies are merely estimates of the degree to which a panelist's knowledge contributes to the consensus view and is related to the probability that he or she would correctly answer any question drawn from the same knowledge domain. This window displays the Kmap image of panelist competencies. The panelists are represented on this image with identifiers assigned by the consensus-based knowledge validation tool.

Depending on his or her role in the panel, the local panelist may be provided or denied access to the results of the entire analysis including the identities of other panelists. The Kmap window also provides an interface through which the local user can display statistical results in the form of panel and knowledge domain profiles. Internally, each profile is represented as an XML document that conforms to the XML schema definition of the consensus-based knowledge validation tool.

The panel profile contains the competency estimates for all panelists, and the knowledge domain profile gives the knowledge validation metric the ratio of the first two eigenvalues, as characterized above, the consensus knowledge model and other statistics useful for assessing the importance of certain items for consensus derivation and knowledge validation. Depending on the measurement scale of the response data, these other statistics may include correlations between the set of panelists' responses for an item and their set of competencies. The objects returned by the consensus-based knowledge validation system, particularly the knowledge map KMAP , are crucial to the notion of knowledge-based collaboration.

By giving panelists greater insight into the manner in which knowledge is distributed among themselves, the consensus-based knowledge validation system motivates further collaboration and the formation of advice networks. One might also wish to use information about other panelists represented on the map to determine those whose perspective seems most different from their own, then initiate further collaboration in attempt to resolve or explain these differences. The map might also reveal novel thinkers, those plotted apart from others or with negative competency estimates, with whom one might want to further collaborate to determine whether these individuals have new knowledge or insights that would benefit others on the panel.

The map and knowledge saliency metric can also detect the existence of strong biases within a panel. Any of these insights gained from information provided by the SKO object could promote collaboration and contribute towards evolving consensus. The present invention encourages knowledge-based collaboration as follows. To discover collaboration tools that are locally available and used by panelists, the SKO or SKO Wrapper if the SKO is implemented as an XML document requires a client modeling tool to provide a Java object that implements a Java interface, called KmapClient, defined by the preferred embodiment of the present invention.

This interface defines a set of Java methods that the SKO can invoke to query for the names of available collaboration tools and to make a request to initiate collaboration with a certain user of the named tool. The advantage of having individual modeling tools to implement the KmapClient interface is two-fold. First, since each modeling tool has the first-hand knowledge of what collaboration tools are being provided to its panelists, the collaboration tools made accessible through the SKO can be exactly the same as those currently in use.

This eliminates the need for users to learn and use new tools when collaborating through the interface of the Kmap window, as described shortly. Second, the SKO can discover locally available collaboration tools in a consistent and tool-independent manner, which greatly increases its interoperability with a wide variety of tools. As described, the Groove tool is used to provide a suite of collaboration tools to users. In one embodiment of the present invention a KmapClient object has been designed to integrate with Groove collaboration tools as depicted in FIG.

Specifically, this object implements the KmapClient interface on one hand and some application logic to invoke Groove tools per user request on the other. As shown in FIG. When the user clicks on a tool name, the SKO object notifies the KmapClient to start the corresponding tool for the local user and remote user associated with the selected panelist identifier. In another embodiment of the present invention, the consensus model, panel profile and associated metrics available as XML documents in the SKO are made accessible through an application programming interface API so that client applications can programmatically access specific information contained in the SKO.

This allows each client model tool to incorporate and render SKO data in a custom manner that best suits its needs. KmapWrapper provides a user interface for communication and collaboration among panelists. The system and method of the present invention is capable of exception handling and analytical diagnostics. In a first phase of data validation, XML schemas are used to validate input response sets, and the tool checks to make sure that certain data input parameters, e.

In a second phase of data validation, the Java algorithms test the data for completeness and identify places where data is missing. Finally, in a third phase of data validation, errors that occur during the statistical processing of the response data set in the R environment are trapped and presented to the user for resolution. For response data measured on interval and ratio scales, t more sophisticated methods than weighted averages can be used for deriving consensus models such as simulation approaches to computing distribution-free estimates [By showing more precisely which kinds of knowledge accounts most for these differences and how, through further collaboration, these differences dissolve as a consensus evolves.

Consequently, in an additional embodiment the statistical algorithms have been modified to incorporate data augmentation and imputation techniques that enable the derivation of models from incomplete data thereby enabling the consensus-based knowledge validation tool to always compute a consensus model from the most current data available to one of its clients. This view should motivate panelists to use collaboration tools in their IT environment to exchange ideas and, when appropriate, revise their opinions. This form of knowledge-building, and the role played by collaboration and consensus-building, can actually be monitored by longitudinal analysis of KMaps. The Schemer system performs longitudinal analysis on a series of KMaps to compute visualizations and metrics useful for assessing the amount of consensus formation and knowledge-building produced by collaboration.

However, longitudinal analysis is complicated by the fact that the above described MDS algorithm produces a KMap whose axes orientation and scale is arbitrary. This means that before successive KMaps can be compared, and metrics computed, all KMaps used for longitudinal analysis must be referenced to the same coordinate configuration. It is often used to compare ordination results, such as the different point configurations in KMaps computed by the Schemer method and system. In a typical Procrustes rotation, the configurations are re-scaled to a common size and jointly centered, and, if necessary, mirror reflected so that their orientation is coincident. In order to find the optimal superimposition, one configuration is kept fixed as a reference, while the other is rotated successively until the sum-of-the-squared residuals between corresponding coordinates in both configurations is minimized.

Greater concordance between data sets after rotation produces a smaller residual sum of squared differences in Euclidean multivariate space. The R function protest computes Corr, then conducts a randomization test to estimate its significance or p-value by calling the procrustes function repeatedly 1, times , keeping track of the proportion of times the value of Corr obtained for the permuted data is greater than or equal to the observed value.

Along with the rotated plots and correlation between each rotated plot and its reference configuration, a Compactness metric, measuring the overall knowledge variability amongst panelists, is also reported. Based on intra-configuration standard deviation, it is computed as follows:. The value of the metric Compactness approached zero as the configuration becomes more compact, indicating greater consensus amongst panelists. When the user clicks on a tool name, the SKO object notifies the KmapClient to start the corresponding tool for the local user and remote user associated with the selected panelist identifier.

In another embodiment of the present invention, the consensus model, panel profile and associated metrics available as XML documents in the SKO are made accessible through an application programming interface API so that client applications can programmatically access specific information contained in the SKO. This allows each client model tool to incorporate and render SKO data in a custom manner that best suits its needs. KmapWrapper provides a user interface for communication and collaboration among panelists. The system and method of the present invention is capable of exception handling and analytical diagnostics. In a first phase of data validation, XML schemas are used to validate input response sets, and the tool checks to make sure that certain data input parameters, e.

In a second phase of data validation, the Java algorithms test the data for completeness and identify places where data is missing. Finally, in a third phase of data validation, errors that occur during the statistical processing of the response data set in the R environment are trapped and presented to the user for resolution. For response data measured on interval and ratio scales, t more sophisticated methods than weighted averages can be used for deriving consensus models such as simulation approaches to computing distribution-free estimates [By showing more precisely which kinds of knowledge accounts most for these differences and how, through further collaboration, these differences dissolve as a consensus evolves. Consequently, in an additional embodiment the statistical algorithms have been modified to incorporate data augmentation and imputation techniques that enable the derivation of models from incomplete data thereby enabling the consensus-based knowledge validation tool to always compute a consensus model from the most current data available to one of its clients.

This view should motivate panelists to use collaboration tools in their IT environment to exchange ideas and, when appropriate, revise their opinions. This form of knowledge-building, and the role played by collaboration and consensus-building, can actually be monitored by longitudinal analysis of KMaps. The Schemer system performs longitudinal analysis on a series of KMaps to compute visualizations and metrics useful for assessing the amount of consensus formation and knowledge-building produced by collaboration.

However, longitudinal analysis is complicated by the fact that the above described MDS algorithm produces a KMap whose axes orientation and scale is arbitrary. This means that before successive KMaps can be compared, and metrics computed, all KMaps used for longitudinal analysis must be referenced to the same coordinate configuration. It is often used to compare ordination results, such as the different point configurations in KMaps computed by the Schemer method and system.

In a typical Procrustes rotation, the configurations are re-scaled to a common size and jointly centered, and, if necessary, mirror reflected so that their orientation is coincident. In order to find the optimal superimposition, one configuration is kept fixed as a reference, while the other is rotated successively until the sum-of-the-squared residuals between corresponding coordinates in both configurations is minimized. Greater concordance between data sets after rotation produces a smaller residual sum of squared differences in Euclidean multivariate space.

The R function protest computes Corr, then conducts a randomization test to estimate its significance or p-value by calling the procrustes function repeatedly 1, times , keeping track of the proportion of times the value of Corr obtained for the permuted data is greater than or equal to the observed value. Along with the rotated plots and correlation between each rotated plot and its reference configuration, a Compactness metric, measuring the overall knowledge variability amongst panelists, is also reported.

Based on intra-configuration standard deviation, it is computed as follows:. The value of the metric Compactness approached zero as the configuration becomes more compact, indicating greater consensus amongst panelists. In FIG. The individual residuals between homologous points are also interpreted separately in the center column of plots. In these plots, the differences between a panelist's current and preceding location are represented by an arrow, with the head of the arrow pointing to his location in the reference configuration, and the length of the arrow proportional to the residual distance between these locations. The reason for providing all of the visualizations and supporting metrics described above is to foster knowledge-building and intelligence improvement through increased collaboration amongst panelists.

By giving panelists greater insight into how knowledge is distributed within a panel, they will be further motivated to share what they know. This includes a heightened awareness of key concepts and the negotiation of their meaning, and growing agreement on first principles and assumptions key to the relevant domain theory. The present invention provides a means for actually monitoring this process. For example, the plots and statistics computed from a longitudinal analysis of KMaps, presented in FIG. As panelists exchange more information and increase their knowledge of the topic domain, they eventually come to share a similar domain theory, and the following pattern emerges.

This conclusion is further supported by a gradual decrease in the compactness metric over the same time periods. The middle column of plots shows how the knowledge possessed by panelists, with respect to their peers, changes over time. The last plot in this series exhibits relative stability with few panelists having shifted much from their previous position. The rightmost column of plots is a replotting of each KMap after Procrustes rotation, if rotation was applied. The trend in this sequence of plots is for a greater concentration of panelists within higher-valued competency contours. The increase in correlation between successive KMaps in the series also confirms growing consensus and panel convergence on a shared domain theory.

The procedure for making a request for time series analysis and retrieving the generated results, which has been realized in the current embodiment of the present invention, is as follows. Referring to FIG. This request contains a set of request identifiers, each of which corresponds to a prior request for consensus analysis and is referred to as an RI. Upon receiving a request for time series analysis, Schemer WS first retrieves previously generated consensus analysis results from its database as per received RIs.

Then, Schemer WS starts a thread for executing an R script that implements the time series algorithm as previously described. The retrieved consensus analysis results are passed in as input parameters to the script. The successful execution of this script results in a PDF document that contains all the time series analysis results and get stored in the database Upon receiving the RI-TS, the client tool can retrieve the results of a time series analysis request as follows. If the R script has failed to execute, the Schemer WS generates an exception to the client tool If the R script has not yet completed its execution, the Schemer WS also generates an exception to the client tool , so that it can retry at a later time.

Note that results of a time series analysis are captured in a PDF document in the current embodiment of the present invention. However, other formats or technologies for encapsulating and representing these results are also possible. For example, as with SKO XML documents and SKO Wrapper code, time series analysis results can be represented as XML documents, and corresponding wrapper code may be created that renders these results in the client tool as per individual application requirements. The above-described embodiments of our invention are intended to be illustrative only.

Numerous other embodiments may be devised by those skilled in the art without departing from the spirit and scope of our invention. A method for consensus-based validation and analysis of response data gathered from a plurality of panelists in response to a plurality of items regarding a subject matter, the method comprising:. The method of claim 1 further comprising the host device developing a knowledge map from the consensus model. The method of claim 2 further comprising the host device displaying the knowledge map and information about the plurality of panelists on a display. The method of claim 1 wherein collecting the set of response data from the plurality of panelists is performed using at least one collaborative modeling tool.

The method of claim 2 further comprising the host device enabling collaboration between the plurality of panelists using the knowledge map as an interface to one or more collaboration tools. A method for consensus-based validation and analysis of one or more sets of response data gathered from a plurality of panelists in response to a plurality of items regarding a subject matter using at least one collaborative modeling tool, the method comprising:. The method of claim 9 further comprising the host device developing a knowledge map from the consensus model. The method of claim 10 further comprising the host device providing the knowledge map and information about the plurality of panelists for display.

The method of claim 10 further comprising the host device providing an interface to the plurality of panelists through the knowledge map. The method of claim 12 wherein the host device providing the interface further comprises enabling access to one or more collaboration tools through the knowledge map. The method of claim 13 wherein the host device enabling is accomplished by launching an instant messaging session. A system for the validation and analysis of knowledge from a set of response data collected from a plurality of panelists regarding a subject matter using a collaborative modeling tool, the system comprising:.

The system of claim 15 further comprising means for developing a knowledge map from the consensus model. The system of claim 16 further comprising means for displaying the knowledge map and information about the plurality of panelists. The system of claim 15 wherein the means for estimating the empirical point estimate matrix, the means for estimating the competency of the plurality of panelists and the means for computing the consensus model are implemented in software developed in accordance with R programming language.

A method to compare a plurality of knowledge maps derived from a consensus model generated from a set of response data collected from a plurality of panelists using a collaborative modeling tool over a period of time, the method comprising:. The method of claim 19 wherein the method further comprises the host device calculating a compactness metric indicative of an overall knowledge variability between the plurality of panelists.

The method of claim 20 wherein the host device performing the procrustes analysis further comprises:. The method of claim 23 wherein the host device enabling the plurality of panelists to communicate further comprises presenting the plurality of panelists with a graphical user interface to one or more collaboration tools when an identifier is selected.

The method of claim 25 wherein the host device enabling further comprises presenting the plurality of panelists with a graphical user interface to one or more collaboration tools after the identifier is selected. The method of claim 25, wherein the plurality of panelists are at least compared in terms of estimated competencies and relative differences in the domain knowledge of the plurality of panelists.

The system of claim 28, wherein the application is further configured to receive requests to retrieve the results of the script. The system of claim 28, wherein the results of the script comprise at least profiles for the plurality of panelists including competency measurements and knowledge domains including consensus values computed for an instrument. The system of claim 28, wherein the statistical programming element is further configured to derive a knowledge map comprising a graphical display comparing the plurality of panelists based on at least the estimated competencies of and relative differences between the plurality of panelists. A tangible computer readable medium having instructions stored thereon, that if executed by a device, cause the device to perform a method for consensus-based validation and analysis of response data gathered from a plurality of panelists in response to a plurality of items regarding a subject matter, the method comprising:.

The tangible computer readable medium of claim 32 having instructions stored thereon, that if executed by a device, cause the device to perform a method, the method further comprising developing a knowledge map from the consensus model. The tangible computer readable medium of claim 32 having instructions stored thereon, that if executed by a device, cause the device to perform a method, the method further comprising providing the knowledge map and information about the plurality of panelists for display. The tangible computer readable medium of claim 32 having instructions stored thereon, that if executed by a device, cause the device to perform a method, the method further comprising providing an interface to the plurality of panelists through the knowledge map.

The tangible computer readable medium of claim 35 having instructions stored thereon, that if executed by a device, cause the device to perform a method, wherein providing the interface further comprises enabling access to one or more collaboration tools through the knowledge map. The tangible computer readable medium of claim 36 having instructions stored thereon, that if executed by a device, cause the device to perform a method, wherein enabling is accomplished by launching an instant messaging session. A method for providing consensus-based validation and analysis of response data gathered from a plurality of panelists in response to a plurality of items regarding a subject matter in web service definition language, the method comprising:.

The method of claim 38, wherein the applet executing further comprises providing an interface to the plurality of panelists through the knowledge map. The method of claim 39, wherein providing the interface further comprises enabling access to one or more collaboration tools through the knowledge map. The method of claim 40, wherein enabling is accomplished by launching an instant messaging session. Justia Patents Modeling By Mathematical Expression US Patent for System and method for consensus-based knowledge validation, analysis and collaboration Patent Patent 7,, System and method for consensus-based knowledge validation, analysis and collaboration.

Justia Patents Modeling By Mathematical Expression US Patent Consensus In Knowledge Analysis System walter elias disney method for consensus-based knowledge validation, analysis and Consensus In Knowledge Analysis Patent Patent 7, Consensus In Knowledge Analysis and method for consensus-based One Characters Craziness In Hamlet validation, Consensus In Knowledge Analysis and collaboration. The present Consensus In Knowledge Analysis encourages knowledge-based What Is Machu Picchu? as follows. In addition, a knowledge map KMap is included, which is a contour image Consensus In Knowledge Analysis graphically displays relative Consensus In Knowledge Analysis of the panelists in terms of their Consensus In Knowledge Analysis competencies and relative differences in their cask of amontillado knowledge as depicted in Kmap window in FIG. Ann Math Stat — Consensus In Knowledge Analysis step the response data Consensus In Knowledge Analysis by a collaborative modeling Consensus In Knowledge Analysis or tools such as the SIAM tool is organized into Consensus In Knowledge Analysis formal model Consensus In Knowledge Analysis a response data matrix X containing the responses X ik of Consensus In Knowledge Analysis l. In Consensus In Knowledge Analysis preferred embodiment Barrier In Nursing the present invention, the Consensus In Knowledge Analysis and method is implemented as a Consensus In Knowledge Analysis service.

Current Viewers: