Computer Decision Making Technologies

Abstract

We interact with computers and computer-based systems in our everyday life. We have them installed virtually in all working places and carry them around with us. In our interactions with these powerful devices, we solve complicated problems, calculate, process and analyze data, create art and communicate with others (Marko, 2009). Advanced computer technologies and innovations have gone a notch higher today. Technological innovations and the development of computer-based systems such as artificial intelligence, big data, cloud computing, machine learning, quantum computation, and decision support systems have come a long way to support computer users in their daily work, and have found their place in almost all industries- meteorological department scientific research, medicine, finance, military, disaster management, agriculture, aerospace engineering, and learning among many other industries (Nada, Marko, Moyle, & Mladenic, 2012).

This paper focuses application of computers in decision making processes, and looks at how technologies such as big data, artificial intelligence, binary code, and quantum computations help in decision making in computers. Through both quantitative and qualitative analysis, the papers answer these questions, gives a clear picture of where we are technologically headed, and at the end are conclusions and recommendations based on the research’s outcome.

Part One: Primary Solution Analysis

Introduction and Definitions

Decision making is a vital activity in human life since it helps determine the course of actions that people find fit, both at individual, group and organizational capacities. Human brain may ‘carry’ high-level information, but unfortunately is not a reliable ‘tool’ in decision making. Thus, for computer-based systems and technologies help analyze such complex information and data. Additionally, computers can process more information faster and store many complicated data in their ‘brains’. Computer generated decisions are more accurate and faster when compared to human brain hence increased overall returns. Technologies such as Big Data, artificial intelligence, binary code and quantum computations have over the years helped improved computer-based decision-making processes.

Essay writing service:

• Excellent quality
• 100% Turnitin-safe
• Affordable prices

Order now

Big Data refers to scientific processes, driven by specialized analytics systems and software, through which large and varied data sets are examined in a bid to uncover patterns, correlations, market trends, customer preferences and behavior, and other information that help organizations position themselves on profitability paths. These processes help understand new revenue opportunities, develop their strategic plans, better their customer services and improve on competitive advantage and operational efficiency (Ram , 2016). Research shows that organizations that have incorporated Big Data analytics technologies have realized an over 26% increase in performance efficiency in the past three years; and are expected to further improve by 41% in the next three years too (Paul, 2012). This is made possible by the powerful capabilities of big data analytics that help business leaders and management have a better view of business problems, decide faster, react quickly to emerging issues, and make ‘bulletproof’ decisions (UKISUG, 2017).

Big Data has been in existence since 1663 and started when John Graunt while studying the bubonic plague that was ravaging Europe, dealt with overwhelmingly huge and uncontrollable data to understand the root cause of the plague. The concept has since been advanced until in the recent past when it has become more popular thanks to the development of computer technologies and more advanced analytical algorithms (Keith, 2017).

Get a brand-new guide on essay writing

Everything you need to get A+ this semester!

Get it

By continuing, you agree to the Terms And conditions and Privacy policy

The download will start within seconds. Good luck!

Download

Artificial intelligence is another technology through which computer decision making processes are being aided. By definition, AI refers to a process through which machines are given the ability to replicate and/or exhibit cognitive human features and characteristics. AI has improved computer-decision making process through the use of intelligent and more efficient algorithms and agents, also called bots. Artificial intelligence-based system feeds on a large amount of data (Big Data) and processes it to produce highly sophisticated models upon which managers base their decisions.

Quantum computing: after decades of being used to binary based info processing computers, the world yet again takes a shift to and focuses now in the development of quantum computers, and that which are expected to be more powerful than the current ones, in terms of their processing and computational power. Instead of using bits and switches that equal to either 0 or 1, quantum computers use qubits- superstitions of 0 and 1 during calculations. A qubit is thought of asset of coordinates used to define one unit long vector. What makes quantum computing more important in the present day world, though at its infancy stages, is its powerful capabilities of solving security and encryption related issues (Perlner & David, 2009) and complex decision-making.

Get your paper done on time by an expert in your field.

Learn more

With the introduction of quantum computing, though at its infancy age, is an idea that the world must be ready to embrace as it offers many promises, just like Big Data. Research has it that we are bound to have an influx of mainstream computing machines that will have high variability in terms of architecture, capacities, and capabilities. It is evident that future computer, communications, and network systems will be a combination of classical and quantum technologies. Additionally, artificial intelligence presents greater advancements through robotics engineering. In the near future, we could have robots that will look, feel and to a greater extent act like human. These robots are expected to accomplish more and complicated tasks, and make decisions by help of highly sophisticated software that will place pattern recognition at the core of the process rather than artificial intelligence.

Part Two: Compilation of Research, Data, and Other Pertinent Information

Big Data Analytics Models

Big data analytics take four forms- irrespective of the field in which they are used, and are applied to manage patient population data in health industry, consumer sentiments analysis, complex competitive analysis, and managing smart power grids among many other application areas.

Prescriptive business intelligence refers to a type of analysis which though not used in most cases, sheds a lot of light on a subject by giving a laser-like focus to answer specific business questions. Predictive analytics use big data to identify past data patterns thuds helping in predicting the future. This analytics target social media communications, documentation, and CRM-related data. In some cases, predictive data analytics models have been used to support sales and marketing operations and other complex business forecasts. Diagnostic analytics are meant to reveal or explain why something happened while descriptive analytics- also called data mining- are used to uncover important patterns in data that could offer great insights. Descriptive analytics have been successfully deployed in credit risks assessments and sales cycle studies. The chart below illustrates the increasing popularity of big data analytics over the past few years. It is evident that Big Data is fast taking lead in enterprise computations and decision-making processes.

Figure 1: increasing big data popularity

Figure 2: big data analytics process

Decision making in computer systems can be looked into as two stage process with data management and analysis processes as the only phases. During data management process, appropriate data is acquired, recorded, extracted, cleaned and integrated. The analytics stage uses modelling techniques to analyze and interpret data.

Part 3: Problem Solving Procedure

Decision Questions in Big Data Analysis

Data analysis using computer technologies depends on input variables which to some great extent, limit the integration and decision modeling process too. Decision models’ variables originate from multiple sources- internal or external, hence it is important to ascertain these sources as a way of maintaining semantic interoperability. Managerial decision will require input data from economic indices, financial data, time series, inflation rate, assets value estimations, and prediction estimations. Alike, decision models in the field of medicine will take symptoms, biochemical analysis results data and anamnestic data as inputs. The process must answer paramount questions in relation to the five features of Big Data- variety, veracity, velocity, volatility, and volume.

1. Volume- how much data do the researchers need? Is it simplified, detailed, sampled or summarized? In social media sentiment analysis, for example, actual tweets or media message exchanges is considered.
2. Volatility- this question answers questions regarding expected changes in data that is being used to formulate decision questions. It advises on how fast data must be utilized in order to provide value. The surest way of ensuring that data changes don’t affect your analysis is by sampling data at different time intervals during the day.
3. Velocity- how fast is the data being recorded? Is it in real-time, daily or monthly? Once this question is answered, samples of such data can be taken, sampled to obtain data feed and pattern prediction from historical data where data is not in real-time. At times, depending on the type of industry, geotagged data is an important target.
4. Veracity- this quality describes data in terms of accuracy, how much can the sources be trusted to answer decision questions at hand? Is the data we are relying on validated, verified, and is its variability known? For example, in Twitter message analysis, Mitchell et al. (2003) argue that although demographics are difficult to ascertain, 15% of online adults use Twitter, with the majority of tweeters living in the urban and that 18-29 years old and the minority are highly represented on Twitter compared to other populations.
5. Variety- this is the very first question that big data analysts must answer as it carries much weight, and answers to the above questions/assumptions depend on it. The considerations made here are whether data to be analyzed is either internal or external, transactional, social media, or web-generated. As an example, a company that invested in Twitter feeds will in addition to internal transactional data, target social media-generated data in response to loyalty sales and marketing department.

Real Case Study: sentiment analysis of Twitter data with keywords hotel and comfort in hotel and tourism industry.

Below is an analysis of Twitter-generated data for Choice Hotel

 Table 1: Twitter sentiments Analysis data (Phillips, Gloria, & Angela, 2014)

The above-tabulated data is descriptive data collected analyzed from tweets sent within 10 minutes following a holiday weekend. Visual analysis indicated that the tweets were from individuals and not businesses. From the study, it can be seen that Twitter data can be used to provide marketing insights and that a variety of information can be obtained from such analysis, and which could be used as foundations or basis for important business decisions. A marketing strategy, for example, could be started as a way of selling the hotel brand as a center for comfort. However, such analysis would require investment in human resources- business analysts equipped with knowledge and expertise to develop meaningful computer analytics programs for the industry.

Part Four: Recommendations, and Conclusions.

Recommendations

Big Data presents big opportunities for businesses in the present day world. However, there are some few technical, strategic, organizational and institutional challenges that must be addressed before the whole idea takes roots in individual organizations. Inconsistent and unstandardized data, finding the Return On investments in data analytics, employing staff with hands-on skills for big data analytics, implementing big data analytics systems and training people to use them, aligning data-driven decisions with business strategic plans, and obtaining a global view of data that is relevant to the organization are some of the issues that must be addressed as far as the technology is concerned.

Applications of computer and computer systems in decision making is a vast growing idea. With the introduction of quantum computing, though at its infancy age, is an idea that the world must be ready to embrace as it offers many promises, just like Big Data. Research has it that we are bound to have an influx of mainstream computing machines that will have high variability in terms of architecture, capacities, and capabilities. It is evident that future computer, communications, and network systems will be a combination of classical and quantum technologies.

We can write
your paper for you

100% original

24/7 service

50+ subjects

Hire a Writer

Conclusions

Better decisions lead to provision of improved information. The analysis of decisions modelled by computer based systems starts from the basic assumption that humans and their brains are bound to be limited, or would rather depend on their intuitions, suspicions and feelings when making decisions. Computers are efficient devices to be relied on when making critical decisions in the world of business. With such emerging technologies as big Data, quantum computing, machine learning, and artificial intelligence, decision making and strategic planning processes much more information faster than a typical human brain can. Big data analysis is characterized by huge volumes, high velocity, veracity, and variability and variety of sources of data, features that keep it at the center of decisions making processes in most organization.

Appendix 1

Definition of terms

CRM- customer relationship management system

Appendix 2

Real case study: Choice Hotels

The case study used in this research refers to big data analytics for a hotel franchisor in the United States. The hotel has over 6500 location in US and 35 in other countries. Its portfolio includes comfort suites and it forms one of the largest limited service brands in the United States.

Appendix 3

Sample Tweets Mining Code

/*searching for tweets with the word hotel or comfort*/

tweets[‘hotel’] = tweets[‘text’].apply(lambda tweet: word_in_text(‘hotel’, tweet))

tweets[‘comfort’’] = tweets[‘text’].apply(lambda tweet: word_in_text(‘comfort’, tweet))

/*checking the relevance of each tweet*/

tweets[‘relevant’] = tweets[‘text’].apply(lambda tweet: word_in_text(‘programming’, tweet) or word_in_text(‘tutorial’, tweet));

/*counting and printing these tweets*/

print tweets[‘hotel’].value_counts()[True]

print tweets[‘comfort’].value_counts()[True]

print tweets[‘relevant’].value_counts()[True]

Decision making using Artificial Intelligence

S_t+1R_t

Above is a basic representation of decision making process using artificial intelligence. At the very basic level, the agent acts on its environment and receives feedback on the effects of any action, and finally selects and executes actions in a successive manner until a predefined condition I achieved.

Reinforcement learning achieves automated optimal decision by addressing portfolio optimization algorithm called Q-learning algorithm. The algorithm builds on dynamic programming and its equations take the form of bellman optimality Equations:

Q* (s, a) = E {r_t+1 + y max _a’ Q*(s_t+1, a’) |s _t = s_, a _t = a} and  P_ss^a [R^a _ss’ +y max_a Q*(s’, a’)] , where y is discount factor for the future reward, Q*(s, a) represent the optimal action (a) value that maximizes or minimizes immediate reward in state (s).