A software system for intelligent mathematical optimization of content marketing. Utku Kose. Computer Sciences App. and Res. Center. Usak University. Usak ...
A software system for intelligent mathematical optimization of content marketing Utku Kose
Ahmet Demir
Computer Sciences App. and Res. Center Usak University Usak, Turkey
Karahalli Voc. School - Dept. of Foreign Trade Usak University Usak, Turkey
Hasan Armutlu
Selcuk Sert
Karahalli Voc. School - Dept. of Computer Tech. Usak University Usak, Turkey
Karahalli Voc. School - Dept. of Foreign Trade Usak University Usak, Turkey
Abstract marketing approaches employed by companies. Because of its connections with especially social media, it is always important to obtain effective content marketing processes in the context of a dynamic, flexible communication environment. So, there is a remarkable research interest in making everything better for content marketing. In this paper, it is aimed to develop a software system, which is able to use artificial intelligence for optimizing parameters of a content that may be provided over popular social media environments for marketing purposes. By optimizing the content according to feedbacks from users, it is thought that next presentation of the content may result to improved interest by objective users. Here, it has been tried to be done thanks to a software system. Keywords- content marketing, artificial intelligence, optimization, software system, intelligent optimization
I.
INTRODUCTION
Technological developments have a great role on changing practical tools technologies can easily improve our life standards and open doors to a more tolerable live among all possible troubles in the world. Rise of such technologies has gained momentum after first appearance of especially computer and communication technologies. After the transformation of the society into the informatics society, computer oriented technologies have affected almost all fields of the life. Thanks to the revolutionary communication technology: Internet, the desired information has been more reachable, editable, and sharable. Under all of these changes and developments, it has been more important to process information / data fast and use it immediately for further works. Briefly, we can see effects of such developments in almost every field around us. In this manner, marketing is one of them. As a remarkable field from the perspective of especially business, marketing is an important field, which attracts also researchers. Even the associated literatures have many research works regarding to improving marketing processes by taking some solution ways into consideration. On the other hand,
because this field is highly sensitive to technological developments, it has been improved in time, thanks to especially high interaction and communication possibilities provided by Web for people all around the world. As a result, content marketing, in which entity and delivery operations related to products are done digitally [1-3] has been introduced. With increasing popularity of social media and interactive Web platforms, content marketing has gained importance for companies to achieve their marketing processes by reaching to more people over Web / Internet. At this point, dynamic, interactive and flexible structure of the Web allows designing effective content to promote products effectively over the Web, It is clear that content marketing has enabled companies to design and manage their marketing processes effectively by benefiting from advantages of the digital world. But on the other hand, there has been always a need for making everything better to improve marketing outputs. So, since its first introduction, there has been a remarkable research interest in the context of content marketing field. In the associated literature, it can be clearly seen that there are many different kind of works that were focused on different sides of content marketing. Some of these works can be expressed briefly as follows: In his work, Lieb has focused on how to run marketing in online and social media based environments [4]. Ho and Dempsey have performed a remarkable research on viral marketing, which is a widely-used online content marketing approach [5]. In this context, Harrang and his friends also have a patent regarding to running viral marketing over social media [6]. Vishwanathan and his friends have focused on the way of providing online content marketing to especially mobile device users [7].
In their work, Holliman and Rowley have focused on marketing [8]. In addition to the related works, it has been also a common approach to think about alternative strategies to improve and optimize content marketing as it can be seen in [9-12]. Starting from the idea of achieving better ways of content marketing over online platforms, it is a remarkable approach to consider combination of different technologies to improve experiences. In this context, the authors have thought about an alternative, effective approach running over a software system. Objective of this paper is to develop a software system, which is able to use artificial intelligence for optimizing parameters of a content that may be provided over popular social media environments for marketing purposes. By optimizing the content according to feedbacks from users, it is thought that next presentation of the content may result to improved interest by objective users. Here, it has been tried to be done thanks to a software system. In detail, the software system employs an optimization algorithm (which is swarm intelligence oriented) to optimize the parameters determined by the company experts to evaluate success of their content provided over the Web platforms. Organization of the remaining sections is as follows: The next section focuses on artificial intelligence based optimization and the concept of content marketing. Additionally, this section also provides a brief explanation regarding to optimization solutions that can be performed over content marketing processes. After this section, the third section explains the details of the developed software system and finally, the last section provides conclusions along with a brief discussion about possible future works. II.
INTELLIGENT OPTIMIZATION AND CONTENT MARKETING
The optimization approach considered in this study is associated with the use of artificial intelligence based techniques. In this sense, a possible model of content marketing process will be supported with a software system, which includes artificial intelligence based optimization algorithms to achieve the mechanism of optimization. Before explaining the details regarding to the developed software system, it is better to enable readers to have some essential information about two main actors of the study as artificial intelligence based optimization approach and content marketing: A. Mathematical Intelligent Optimization with Artificial Intelligence Intelligent optimization processes performed with artificial intelligence are generally done thanks to specially designed algorithms. As essential techniques of the artificial intelligence, these algorithms are effective at finding the optimum value by using some kind of intelligent particles that are able to search for optimum value(s) within a given solution space (and according to some constraints). In such algorithms, the
following factors are essential approaches for running mathematical intelligent optimization (even the objective problem discipline is different) [13]: Inspiration from the nature: The nature has an important role on designing and developing intelligent techniques for mathematical optimization. This can be done by simulating dynamics of the nature (or living organisms in the nature) by using logical steps and mathematical approaches. Steps based on iteration / error or a desired value: Artificial intelligence based intelligent optimization approaches are organized over algorithmic structures. Because of that, it is needed to stop running of such algorithms according to an objective value. This objective value can be a number of total iterations (completed turn of the whole algorithm), an error value, or a desired optimum value that can be obtained within the algorithmic steps. Collective particles: Intelligent mechanism behind artificial intelligence for finding the optimum value is associated with movements logically and mathematically done by a group of intelligent particles. These particles are able to move in the solution space by also interacting with other particles moving and updating their next moves according to situations of other particles (or its past situations). Eliminating the worse / improving the better: While searching for optimum value(s) with artificial intelligence techniques, the most common way that is used is to eliminate worse particles (removing them from solution space or changing them with new particles) or improving active particles having better solution parameters (i.e. particles in top five or the best particle so far). In optimization problems, it is also possible to have the appropriate objective function (fitness function) that will be used along the optimization process. This function / equation include the related parameters to be optimized and any other constant values that affect these parameters. In addition, there are some constraints limiting values of the parameters, as it was mentioned before. Some simple structures of an objective function can be in the forms shown as follows [13]: (1) (2) (3) By taking the related objective function into consideration, the artificial intelligence based optimization algorithm then try to find optimum parameter(s) according to a desired value. While finding the optimum minimum value, the problem is some kind of minimization, and while finding the optimum maximum value then it is some kind of maximization.
In the literature there are many different kinds of intelligent optimization algorithms. It is suggested the readers to refer to these algorithms in order to examine technical infrastructure of these algorithms and understand more about what was done in this study. Some of the related algorithms are: Particle Swarm Optimization [14], Ant Colony Optimization [15], Artificial Bee Colony [16], Cuckoo Search [17], Bat Algorithm [18], Firefly Algorithm [19], Bacterial Foraging Optimization Algorithm [20], Intelligent Water Drops Algorithm [21],
Customer conversion, Customer upsell, Lead conversion and nurturing, Passionate subscribers. After focusing on artificial intelligence based optimization and the concept of content marketing, it is now possible to discuss about possible optimization oriented solutions regarding to content marketing. C. Optimization of Content Marketing As general, there can be many different ways of optimizing content marketing by employing the artificial intelligence. All these aspects of such optimization are just related to what will be optimized, what is the optimization objective, what kind of platform is used for cont the readers can refer to a recent work by Kose and Sert [36].
Vortex Optimization Algorithm [22], Cognitive Development Optimization Algorithm [23],
Sert in their works has been employed for the optimization mechanism of the developed
Algorithmic Reasoning Optimization [24], Fish School Search [25], Krill Herd Algorithm [26], Clonal Selection Algorithm [27], B. Content Marketing Content marketing is a trendy marketing way followed by can be defined briefly as a marketing approach, which aims to find products produced
expressed briefly as follows [36]: In the model of optimized scenario, the content marketing is associated with a scenario, which employs some values parameters (i.e. number of total views of the scenario, total number of sold, total costs, income, and outcome) used for calculating an objective success ratio. The related success ratio is based on a general equation that may be formed by expert(s).
and fulfillment in this way [28]. According to the Content Marketing Institute [29] focused on creating and distributing valuable, relevant, and consistent content to attract and retain a clearly-defined audience - and, ultimat
In the equation, there are also some additional variables, which should be optimized (i.e. total number of next further view for the scenario, total number for target customers / users, application priority of the scenario among some other alternatives).
interest to the products of the company [30].
Through the model, the variables are optimized according to objective success ratio and next moves regarding to destiny of the scenario is determined by the people controlling the content marketing process.
Steimle indicates that the key factor in content marketing is value [31]. He says briefly about this s if a piece of content is the sort that could be part of a content marketing campaign if people seek it out, if people want to [31]. From a general perspective, it is possible to said that content marketing can be in many different types of formats including news, video, white papers, e-books, infographics, e-mail newsletters, case studies, podcasts, how to guides, question and answer articles, photos, blogs etc. [31-34], if the content is enough capable of taking Regarding to objectives of content marketing approaches, the literature comes with many explanations. In this context, it is possible to briefly ideas about objectives of content marketing. They define objectives of content marketing as [35]: Brand awareness or reinforcement, Customer service,
The related solution processes within the model are continued until the desired output is achieved. In the optimization approach provided with the software system here, the related values parameters and additional variables defined for the optimized scenario model (under an equation) are mostly associated with social media oriented parameters like number of likes, number of views, number of shares, number of positive / negative comments. Furthermore, it is also possible to determine different parameters specific for a Web platform on which the content marketing is run. But the main optimization In order to understand more about what is done in the software system introduced here, it is better to examine essential using features and functions of the software system and a typical optimization approach that can be achieved thanks to it.
III.
A SOFTWARE SYSTEM FOR OPTIMIZING CONTENT MARKETING
By taking essential approaches indicated under the previous section for optimization of content marketing, it is possible to run software oriented solutions. In this context, the authors have designed and developed a software system for optimizing an active content marketing process (or at least a planned one). More details regarding to the software system has been explained under the following sub-sections:
B. Optimizing Content Marketing with the Software System A typical optimization process over the software system can be done by doing the following tasks: First of all, the main interface of the software system is used for determining how to start defining the optimization problem of the objective content marketing strategy / scenario. This interface is also for viewing the admin interface if desired (Figure 1).
A. Using Features and Functions of the Software System The developed software system in this study is able to give feedback on how to improve effectiveness (for marketing purposes) of the content provided over the Web (especially over social media). At this point, users can benefit from a window-based (Form) environment. In this sense, it is possible to indicate essential using features and functions of the software system as follows: The software system allows users to design their optimization process via two different ways. The first one is defining their content marketing success equation (objective function), thanks to an interface for writing / typing mathematical equations needed. By determining the equation to be employed in optimization process, the software system automatically finds out essential parameters that will be evaluated. Then the system also wants the users to define if there is any constraint associated with the optimization process. The second way to design the optimization process is to define each parameter that will be used for optimization objectives. After defining the parameters, the software system needs a success equation (objective function) associated with the parameters, so the interface for that is automatically viewed. Again, the system then wants the users to define if there is any constraint associated with the optimization process that will be started.
Figure 1. Main interface of the software system.
By choosing different starting points, the users can interact with parameters defining interface or equation writing interface in the next task of forming the optimization approach (Figure 2). While defining parameters or defining the equation, the user is directed to determine further details regarding to the active parameter / equation element (i.e. limits of the parameter / element, its data type, additional constant values).
The software system allows the users to provide current parameter status of an active content to be evaluated in terms of suggestions for future or feed the system to have possible success values for some planned parameters / components of the content that will be provided over the digital environment (i.e. Web). The software system comes with an easy-to-use platform, with functions on directing users on how to use the controls of the system for ensuring an optimization process. It is also possible for advanced users (i.e. experts, artificial intelligence infrastructure designers) to adjust main evaluation parameters of the artificial intelligence mechanism on the background of the software system. This can be done by reaching to admin interface of the software system.
Figure 2. Parameters defining and equation writing interfaces of the software system.
After finishing works on determining the objective function of the marketing scenario / process, it is then possible to indicate the elements associated with the objective social media or Web environment to be evaluated in the context of optimization over the related equation. Here, it is also possible to feed the system with an already obtained data from an active content that is run currently within a marketing scenario. In detail, the data should be stored in a matrix form under the Excel file format (.xls, .xlsx). All of
these tasks are done over the content element organization interface (Figure 3).
Figure 3. Content element organization interface of the software system.
As default, the system provides essential elements associated with definitions made while defining the objective function. So, the user can directly determine objective success values, any constants affecting the output of the objective function, and indicate final points for the optimization process just before passing to evaluation and results interface. By completing content element organization, the user can start the intelligent evaluation / optimization process done over the related scenario, by clicking on the start evaluation button located over the content element organization interface. After the intelligent evaluation / optimization process, the evaluation and results interface of the system is automatically viewed (Figure 4). Over this interface, the user is informed about further suggestions by the software system to improve effectiveness of the current content marketing process / scenario. It is possible for the user to save the optimization work to be able to open it later, or save the results to examine further to improve the marketing scenario. The feedback / evaluation results provided over this interface by the system are essential products of the artificial intelligence approach designed and developed for improving content marketing as under the objective of this study.
Figure 4. Evaluation and results interface of the software system.
IV.
CONCLUSIONS AND FUTURE WORK
In this study, a software system, which is able to optimize a content marketing process according to the determined success parameters, has been introduced briefly. In detail, the study has also provided some information regarding to possible optimization of content marketing models during applying them over a Web platform like a social media environment. Here, the authors have aimed to employ artificial intelligence based optimization to achieve an intelligent optimization of content marketing over a dynamic software system. The software system introduced in the study allows users to determine their parameters for evaluating success rating of the content within marketing operations and run an intelligent analyze period to improve effectiveness of their current content marketing process. The authors believe that integration of such systems to the objective marketing platforms (or at least to their marketing plans) will enable companies to control and improve their content marketing strategies easily and in a practical way. Potential of the developed software system to ensure an active optimized content marketing process have encouraged the authors to plan some future works. In this context, the system will be used over many different social media and etrade platforms with different kinds of content parameters and the obtained results (different feedback types from users, data regarding to i.e. total followers, emoticons used) will be reported. On the other hand, there will be also some additional works to improve the current software system to newer versions by including at least three different artificial intelligence optimization techniques that can be chosen by users with adjustable parameters for each specific technique. Finally, it is also aimed to develop Web based and mobile application based versions of the system. REFERENCES [1]
N. Koiso-Kanttila, Digital content marketing , Journal of Marketing Management, 20(1-2), pp. 45-65, 2004. [2] Understanding digital content marketing Marketing Management, 24(5-6), 517-540, 2008. [3] S. Rao, V. Srivatsala, and V. Suneetha, Optimizing Technical Ecosystem of Digital Marketing . In Artificial Intelligence and Evolutionary Computations in Engineering Systems (pp. 691-703). Springer India, 2016. [4] R. Lieb, Content marketing: Think like a publisher-How to use content to market online and in social media . Que Publishing, 2011. [5] J. Y. Ho and M. Dempsey, Viral marketing: Motivations to forward online content Journal of Business Research, 63(9), 1000-1006, 2010. [6] J. P. Harrang, D. B. Gibbons and J. M. Burnette, and Methods for Distribution of Digital Media Content Utilizing Viral Marketing over Social Networks U.S. Patent Application No. 12/626,231, 2009. [7] K. K. Vishwanathan, P. R. Iyer and R. Sundar, Methods for Marketing Digital Content to Mobile Communication Device Users , U.S. Patent Application No. 11/626,493, 2007. [8] G. Holliman Business to business digital content . Journal of research in interactive marketing, 8(4), 269-293, 2014. [9] D. Chaffey, P. R. Smith eMarketing eXcellence: Planning and optimizing your digital marketing . Routledge, 2012. [10] Understanding digital marketing: marketing strategies for engaging the digital generation . Kogan Page Publishers, 2014. [11] F. Gavat Digital content marketing: storytelling, strategia, engagement. Edizioni Lindau, 2013.
[12] D. Journal of Research in Marketing, 31(2), 192-206, 2014. [13]
[14] [15] [16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
International Virtual Conference on Advanced Scientific Results / Online Scientific Conference 2016 (ScieConf 2016), 131-137, DOI: 10.18638/scieconf.2016.4.1.380, Slovakia, 2016. J. Kennedy, Particle swarm optimization . In Encyclopedia of machine learning (pp. 760-766). Springer US, 2011. M. Dorigo, M. Birattari and T. Stutzle, Ant colony optimization Computational Intelligence Magazine, 1(4), 28-39, 2006. D. Karaboga and B. Basturk, A powerful and efficient algorithm for numerical function optimization: artificial bee colony (ABC) algorithm Journal of Global Optimization, 39(3), 459-471, 2007. Cuckoo se . In Nature & Biologically Inspired Computing, 2009. NaBIC 2009. World Congress on (pp. 210-214). IEEE, 2009. X. S. Yang, A new metaheuristic bat-inspired algorithm . In Nature inspired cooperative strategies for optimization (NICSO 2010) (pp. 6574). Springer Berlin Heidelberg, 2010. Firefly algorithm, stochastic test functions and design optimisation . International Journal of Bio-Inspired Computation, 2(2), 78-84, 2010. S. Das, A. Biswas, S. Dasgupta and A. Abraham, Bacterial foraging optimization algorithm: theoretical foundations, analysis, and applications . In Foundations of Computational Intelligence Volume 3 (pp. 23-55). Springer Berlin Heidelberg, 2009. H. Shah-Hosseini, The intelligent water drops algorithm: a natureinspired swarm-based optimization algorithm . International Journal of Bio-Inspired Computation, 1(1-2), 71-79, 2009. U. Kose and A. Arslan, On the idea of a new artificial intelligence based optimization algorithm inspired from the nature of vortex . BRAIN. Broad Research in Artificial Intelligence and Neuroscience, 5(1-4), 60-66, 2015. U. Kose and A. Arslan, Realizing an optimization approach inspired . BRAIN. Broad Research in Artificial Intelligence and Neuroscience, 6(1-2), 14-21, 2015.
[25] C. J. Bastos Filho, F. B. de Lima Neto, A. J. Lins, A. I. Nascimento and M. P. Lima, Fish school search . In Nature-inspired algorithms for optimisation (pp. 261-277). Springer Berlin Heidelberg, 2009. [26] A. H. Gandomi and A. H. Alavi, new bio-inspired optimization algorithm . Communications in Nonlinear Science and Numerical Simulation, 17(12), 4831-4845, 2012. [27] L. N. De Castro and F. J. Von Zuben, The clonal selection algorithm with engineering applications . In Proceedings of GECCO (Vol. 2000, pp. 36-39), 2000. [28] A. Karkar, Content marketing on the increase of value and confidence , International Journal of Social Sciences and Education Research, 2(1), pp. 334-348, 2016. [29] Content Marketing Institute, What is content marketing? [Online] at http://contentmarketinginstitute.com/what-is-content-marketing/, 2016, accessed November 10th, 2016. [30] E. Ozgen and H. Doymus, A communicative approach about content management as a differentiator factor in social media marketing (In
[31]
[32]
[33]
[34]
[35]
. AJIT-e: Online Academic Journal of Information Technology, 3(10), 2013. What is content marketing? Forbes [Online] at http://www.forbes.com/sites/joshsteimle/2014/09/19/what-is-contentmarketing/#11393b331d70, 2014, accessed November 11st, 2016. J.-P. De Clerck, Content marketing defined: A customer-centric content marketing definition . i-Scoop [Online] at http://www.iscoop.eu/content-marketing/content-marketing-defined-customercentric-content-marketing-definition/, 2016, accessed November 11st, 2016. D. Farnworth, What is content marketing? Copyblogger [Online] at http://www.copyblogger.com/content-marketing-codex/, 2015, accessed November 12nd, 2016. Omedia24. 2014 Trends in Online Marketing: Content Marketing (In German: Trends 2014 im Online Marketing: Content Marketing) [Online] at http://www.omedia24.de/blog/trends/trends-2014-im-onlinemarketing-content-marketing/, 2014, accessed November 12nd, 2016. R. Rose and Managing Content Marketing , CMI Books, Cleveland, OH, 2011.
[36]
[24]
, Ecoforum Journal, in press, 2017. press, 2016-2017.