COIS13013 Business Intelligence

This assessment item relates to course learning outcomes numbers 3, 4 and 5 as stated on page 1 of the course profile.What is the importance of a collaborative technology Provide the names of three software that are used for collaboration and discuss how this technology is used How is big data different from big data analytics Give three examples where big data analytics is very useful for supporting businesses.

Read the resource material available on the course website about WEKA (in Week 5 and Week 6). Download WEKA software and install it onto your computer (ensure the bundled Java Runtime Environment, i.e. jre is installed). After successful installation of the program, classify diabetes from the data diabetes.arff using J48 classifier (keep other parameters as default). The diabetes.arff file can be found on the course website. Write an analysis report based on the classification results. You need to include the results in your report.

Review the options and views available to answer the following questions:

1. What sort of information is provided by the dashboard What visual objects are used and how are they useful for comparison purposes What quantitative or qualitative measures are being reported
2. Provide a critique of the dashboard design and ways in which the design could be improved. 

Collaborative technologies have been employed in organizations for the major goal of enabling employees with complementary skills to interact, share knowledge and create an understanding that would have been difficult to achieve individually. The collaborative technologies are important in communication within and outside the organization, planning and co-ordination of business activities and carrying out effective and efficient transactions.

Lotus Notes is a popular collaborative technology used by knowledge workers. The software is used in universities to capture, amass and share information through the university wide local are network (Neilson, R.E., 1997, 11).  Both the students and faculty members install the software; the students can easily search users and hence the top-down approach is eliminated in the learning process.

Electronic calendars, for example Google Calendar, create a shared flow of future events and schedules.  Unlike paper diaries and calendars, electronic ones can be shared with many people. The ability to reach a huge number of people makes this software a great tool for heads of departments and supervisors who are expected to track the activities of their supervisees (Payne, S.J., 1993, 34).

Cloud based storage applications like Google drive and Dropbox are used in file sharing and synchronization by a wide range of professionals. Writers are able to do collaborative writing through shared folders. Content creators share their work easily with their audience through well designed and control features that are incorporated in the cloud storage applications.

Big data refers to the complex data sets collected from internal business operations as well as other external platforms such as social media; the huge and complex data is believed to have hidden insights that can be useful in the day-to-day or strategic decision-making. Big data poses a big challenge of storage, cleaning, visualization and finally analysis because of the unstructured and noisy format the data is collected.

On the other hand, big data analytics refer to the methods, techniques and tools employed in the data collection, storage and analysis process. The methods are borrowed from statistical and artificial intelligence theories.  Therefore big data analytics is the combination of both the big

data and the methodologies and tools for extracting insights that can support businesses (Chen, H., Chiang, R.H. and Storey, V.C., 2012,1166).

Marketing departments use social media analytics to understand the needs and preferences of potential and existing clients. By collecting and analyzing profiles, tweets and status updates, marketers are able to gain insights that can enhance the conversations between the organization and its audience.

Governments have employed data analytics to design campaigns and encourage political participation. Policy makers are able to identify target publics to discuss policies through multi-media platforms. Political aspirants analyze into big data to identify and profile their supported for better and targeted messages. Big data has also helped to identify who are more likely to donate and even attend political events.

In the health sector, a huge amount of data is generated from patients’ records over a long period of time. The data is used for understanding the current challenges facing the health sector as well as predicting the unforeseeable problems. Big data has also been used in research of new drugs through simulation and analysis of large amount of scientific data (Raghupathi, W. and Raghupathi, V., 2014, 2401).

Classifications Results

The ‘Weka’ toolkit provides an interface that visualizes and then classifies according to the 9 attributes given as the input. The ‘J48’ algorithm tree- based classifier is used to detect if a certain combination of attributes test positive or negative, that is if diabetes is present or absent. The decision tree has nodes that represent each attribute and a branch that stand for the outcome of the test. When the node and the branch are combined, the leaf gives a yes or no label.

The training of data looks for the attributes, which has data instances with the same value as the target data. This kind of attributes are said to have the highest informational gain. The dataset is iterated to reduce the ambiguity between the target and input values.

In this case, the classifier achieves an accuracy of 73%. It can correctly predict 567 of whether a patient has diabetes or does not have given a certain attributes about the patient. The attributes include age, the body mass index, blood pressure, skin thickness and insulin levels.  

407 instances correctly tested as negative while 108 instances were incorrectly classified as positive. On the other hand, 160 instances were correctly classified as positive while 93 were incorrectly classified as negative.

In conclusion, the Weka software is a powerful tool that can be used to solve the problems of classification; clustering and pattern recognition if given a large dataset that could be related. They are other algorithms, which could complement the J48 method.

The information board provides a clear list of disease type through the  ‘Select an Analysis’ section. The list serves as the entry point to the rest of the information. Once the user gets to choose the disease type, the board displays an overview of the information about the patients ailing from that disease.

The map of the regions the patients comes from is presented in the overview.  Apart from the location, the age brackets and gender of the patient is presented in form of a well-formatted table. In addition to age and gender, the table contains the average years of ailment for each age group. All this three key information are presented neatly in one table. At the lower section, there are graphs that show the relationship between the average length of stay and the hospitalization rate. The line plot shows the negative of relationship between the two variables.

The dashboard employs visual tools like the graph to show relationship and the ‘core process measure compliance ’ graphic to compare the national average with the other figures. The maps also visualize the high prevalence of a disease in one region compared to the other. The geographic location of the patients is qualitative information while the length of stay and the hospitalization rate represents the quantitative information.

Overall the information dashboard is well - designed because it presents a number of variables in a concise and easy to understand format. The use of graphics is recommendable because it enhances the users ability to interpret the information. Good-looking graphics are known to arouse the curiosity of the user (Galitz, W.O., 2007,20). However the design information could be improved by incorporating interactive features where the user can plot histograms to visualize the age groups and the different statistics related to that age group. Do-it-yourself features enhance engagement and memorability of the information. The interaction could be achieved through a drag-and –drop interface that draws different kind of charts and tables. A variety of graphics showing the relationship between the different diseases will provide the users with more comprehensive analysis (Reeves, L.M., Lai, 2004,16).  

References

Neilson, R.E., 1997. Collaborative technologies and organizational learning. Igi Global.

Payne, S.J., 1993. Understanding calendar use. Human-Computer Interaction, 8(2), pp.83-100.

Chen, H., Chiang, R.H. and Storey, V.C., 2012. Business intelligence and analytics: From big data to big impact. MIS quarterly, 36(4), pp.1165-1188.

Raghupathi, W. and Raghupathi, V., 2014. Big data analytics in healthcare: promise and potential. Health Information Science and Systems, 2(1), p.3.

Galitz, W.O., 2007. The essential guide to user interface design: an introduction to GUI design principles and techniques. John Wiley & Sons.

Reeves, L.M., Lai, J., Larson, J.A., Oviatt, S., Balaji, T.S., Buisine, S., Collings, P., Cohen, P., Kraal, B., Martin, J.C. and McTear, M., 2004. Guidelines for multimodal user interface design. Communications of the ACM, 47(1), pp.57-59.