Importance of the Use of Analytics in Requirements Engineering Marina Pincuka Institute of Applied Computer Systems, Faculty of Computer Science and Information Tech- nology, Riga Technical University, Kalku 1, LV – 1658, Riga, Latvia marina.pincuka@rtu.lv Abstract. Requirements Engineering is regarded as one of the most important functions in software development process. Inadequate/ incorrect engineering of requirements may lead to expensive errors in software development or even to project failure. Even though there are a different methods and approaches that are proposed in literatures, many of these approaches have not been used in the in- dustry or have been proved to be ineffective. The main goal of this work is to investigate the Requirements Engineering weak points and see which of these weak points can be strengthened by the use of analytics. Keywords: Requirements Engineering, Analytics, Requirements Engineering Challenges 1 Introduction In project, requirements definition is an important function that affects project pro- cesses and results. Dissatisfaction of the requirements may lead to unhappy customers, incorrect system processes or even project failure. The use of analytics allows to pro- cess the information that otherwise may be ignored or overlooked. Based on Dick J., Hull E. & Jackson K. [3], Requirements Engineering (RE) is the subset of systems engineering concerned with discovering, developing, tracing, analys- ing, qualifying, communicating and managing requirements that define the system at successive levels of abstraction. Dankov Y. and Birov D. [2] describes analytics as the process of developing action- able insights through problem definition and the application of statistical models and analysis against existing and/ or simulated future data. The purpose of this paper is to reflect the results of the research in progress con- cerning the importance of the use of analytics in Requirements Engineering functions. Paper reports on three research questions: (i) “What is the state of art of the use of analytics?” (ii) “What are the Requirements Engineering challenges?” and (iii) “How the use of analytics in specified Requirements Engineering function can improve the weak point of Requirements Engineering?”. The paper is structured as follows. The survey of the state of art of the use of analytics is presented in Section 2. Challenges in Requirements Engineering are presented in Section 3. Summary of Requirements Engineering challenges and the use of analytics Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0) 73 to reduce the Requirements Engineering weak points are described in Section 4. Brief conclusions and directions of further research are stated in Section 5. 2 A Survey of the Use of Analytics To understand the state of art of the use of analytics, a literature search was conducted using terms “use” and “analytics”. 25 sources where selected and analysed using Springer Link, Science Direct, IEEE resources, 13 of the sources, where identified as overlapping and are not included in a review. Collected sources of types of analytics are new and do not overlap with previous overview published in Pincuka and Kirikova article “Types of Analytics in Requirements Engineering” [16]. New overview was made to expand understanding of use of analytics in different fields. The gathered articles, where analysed as follows: (i) “What is the industry in which analytics are used?”, (ii) “What kinds of analytics are used?” and (iii) “What the ana- lytics is used for?”. Gathered kinds of analytics and their brief definition is described below: 1. Game analytics – applying analytics and big data in the gaming context [11]; 2. Web analytics – the measurement, collection, analysis and reporting on Inter- net data for the purposes of understanding and optimizing Web usage [1]; 3. Visual analytics – information visualization that focuses in analytical reason- ing facilitated by interactive visual interfaces [14]; 4. Descriptive analytics –describes what is happening or why something hap- pened [19]; 5. Predictive analytics – provides foresight and make predictions about the like- lihood of a future event [19]; 6. Prescriptive analytics – provides support for making decisions, or some cases independent form its own decisions [19]; 7. Business analytics- comprised of solutions used to build analysis models and simulations to create scenarios, understand realities and predict future states [13]; 8. Big Data analytics – the use of advanced analytic techniques against very large diverse data sets that include structured, semi – structured and unstructured data, from different sources and different sizes [13]; 9. Diagnostic analytics – the form of advanced analytics that examines data or content to answer the question, “Why did it happen?”. It is characterized by techniques such as drill-down, data discovery, data mining and correlations [5]; 10. Text analytics – a process of converting unstructured text data into meaningful data for analysis, to measure opinions, reviews, feedback, to provide search facility, sentimental analysis and entity modelling to support fact based deci- sion making [18]; 11. Learning analytics - the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs [4]. 74 Table 1 summarizes all identified types of analytics, industries in which analytics are applied and usage of analytics in different use cases. These analytics are further used in Section 4 in example to overcome Requirements Engineering challenges using ana- lytics. Table 1. Use of analytics in different industries Type of analytics Industry Usage Game analytics Information Technology: Understand use preferences and behaviour Game development [11] [11]; Improve decision – making [11]; Gather insights [11]; Reduce the risk of failures [11] Web analytics Information Technology [1]; Optimizing website functionality and con- Website management [1]; version [1]; Marketing [1] Analysis of past performance [1]; Optimizing performance of and conver- sions from marketing campaigns [1]; Determining the best creative executions through testing [1]; Baseline information for site redesign [1]; Predictive metrics for developing future marketing campaigns [1]; Budgeting and planning for upcoming business objectives [1] Visual analytics Medicine [14] Trend monitoring [14]; Anomaly detection [14]; Testable hypothesis detection [14] Descriptive ana- Medicine [19]; Understand why something happened [19]; lytics Marketing [13] Provide information about past behaviour, patterns or trends in the data [13]; Categorize, characterize, consolidate and classify data to valuable information [15] Predictive Medicine [19]; Prediction of the occurrence of future analytics Information Technology events [19]; [13]; Supports decision – making [13]; Marketing [13]; Understanding of behavioural patterns and Insurance companies [13]; trends [13] Aviation companies [13] Prescriptive ana- Medicine [19] Supports decision – making [19] lytics Business Marketing [13] Prediction of the future states [13]; analytics Understand of the reality [13] Big Data Marketing [13]; Supports decision – making [13]; analytics 75 Type of analytics Industry Usage Human Resource manage- Understanding of behavioural patterns and ment [7]; trends [13]; Construction industry [15];Capturing the strategic linkage [7]; Information Technology [24]Improve performance [7]; Forecast future threats and opportunities [24]; Enhance organizational performance [24] Diagnostic Construction industry [15] Evaluation of the potential causes of a analytics problem [15] Text analytics Information Technology: Risk identification [8]; Mobile application develop- Benefit identification [8]; ment [8] Supports – decision making [18] Learning analyt- Education [4] Understanding and optimization of pat- ics terns [4] The collection of types of analytics and the industries in which they are used allows to (1) understand the main domains of application of analytics and (2) collect different methods and approaches how analytics can be applied in different fields, to understand how analytics can be applied in Requirements Engineering. Gathered analytics are described without any classification into groups, because the purpose of this article is to understand use of analytics in different fields and if it is possible to apply this analytics in Requirements Engineering and overcome Require- ments Engineering challenges. 3 Requirements Engineering Challenges Requirements Engineering deals with a lot of challenges, for example, authors Kahan et al. [9] mentions, previous identified requirements issues/ challenges like - business process focus, systems transparency, integration focus, distributed requirements, layers of requirements, packaged software, centrality of architecture, independent complexity and fluidity of design. Author Schmid [22] mentions Global Requirements Engineering challenges that are based on location of the stakeholders, in his article he mentions challenges with International Requirements Engineering and Distributed Requirements Engineering. In this section Requirements Engineering challenges are collected based on Require- ments Engineering phase in which challenge appears on. Requirements Engineering function has a four phases [12]: • Elicitation is the act to determine or obtain the relevant requirements for the development of a solution. In this phase requirements are identified and dis- covered; • Requirements analysis is a phase, where a user’s requirements should be clar- ified, categorized and documented to generate the corresponding specification. 76 In this phase requirements classification, representation, derivation and nego- tiation are provided; • Requirements specification describes the phase, where the requirements are brought into a suitable and unambiguous form. The idea in this phase is to document the requirements, and to make the requirements document readable and understandable to anyone; • Requirements validation is to review or validation requirements for clarity, consistency and completeness. In this phase requirements faults are identified. Table 2 presents challenges in Requirements Engineering and phases of Require- ments Engineering in which challenges usually emerge. Table 2. Challenges in Requirements Engineering Requirements Engineering phase Challenges Elicitation Incorrect understanding of the requirements [12][6]; System knowledge may be fragmentary, distributed and tacit [21]; Lack of information [23][20][6]; Problems with client and customer representatives [23]; Problems in communication [23][10]; Conflicting requirements [20][10][6]; Random/ uncertain/ unclean requirements [20][10]; Unrealistic requirements [10]; Analysis Integrating physical objects with information objects [21]; Change/ volatility of requirements [23][20][10][6]; Requirements quality issues[23]; Neglect of non-functional requirements [23]; Incomplete requirements [20]; Specification Minimal documentation [23][10]; Complexity of requirements documentation [10]; Validation Requirements validation[23]; Inadequate requirements verification [23]; Summarizing all of the challenges from different scientific papers (see Table 2.) the conclusion can be made, that often authors emphasizes the same challenges in the same phases of Requirements Engineering function. 4 Use of Analytics to Reduce Requirements Engineering Challenges In this Section using summarization of Requirements Engineering challenges are pro- vided the types of analytics, which can be used to improve Requirements Engineering issues. In table 3 are mentioned only those Requirements Engineering challenges, to 77 whom methods of analytics can be applied to, for example, analytics can not be used with challenges in customer/ client representatives or minimal documentation, in these cases other methods should be provided. Analytics, that can be applied to Requirements Engineering challenges where identified in a literature survey, based on analytics use cases and author master thesis Pincuka M. “Analytics in Requirements Engineering” [17]. Analytics, which are proposed to use in Requirements Engineering challenges are chosen based on use of analytics in a literature survey. Table 3. Proposed types of analytics in Requirements Engineering challenges Requirements Engineering challenge Types of analytics Incorrect understanding of the re- Business analytics; Game analytics; Predictive ana- quirements lytics; Prescriptive analytics; Text analytics; Web analyt- ics. System knowledge may be fragmen- Big Data analytics; Descriptive analytics; Learning tary, distributed and tacit analytics; Visual analytics Lack of information Big Data analytics; Descriptive analytics; Text ana- lytics Conflicting requirements Descriptive analytics Random/ uncertain/ unclean require- Descriptive analytics; Diagnostic analytics; Predic- ments tive analytics, Prescriptive analytics Unrealistic requirements Business analytics; Descriptive analytics; Diagnos- tic analytics Change/ volatility of requirements Big Data analytics; Descriptive analytics; Diagnos- tic analytics; Predictive analytics, Prescriptive ana- lytics; Text analytics, Web analytics Requirements quality issues Business analytics; Descriptive analytics; Diagnos- tic analytics Neglect of non-functional require- Big Data analytics; Text analytics ments Incomplete requirements Descriptive analytics; Diagnostic analytics; Predic- tive analytics, Prescriptive analytics Types of analytics that are provided in Table 3, only points out some of the analytics that can be used to improve or overcome Requirements Engineering challenges. For example, (i) business analytics can be used to understand the reality, if we will apply this analytics to Requirements Engineering it will help to better understand the require- ment and it is meaning, (ii) if we identify conflicting requirements using descriptive analytics data can be classified and meaningful information about requirements can be found or (iii) using text analytics with keywords about the system we can identify some additional information to reduce lack of information. To summarize all methods of an- alytics that can be used to overcome Requirements Engineering challenges a literature study must be provided with focus on analytics and use cases, where methods of ana- lytics are used. 78 5 Conclusions In this paper Requirements Engineering challenges are discussed and the use of analyt- ics to overcome these challenges envisioned. The paper contributes (i) a preliminary survey on the use of types of analytics in different industries, (ii) summary of Require- ments Engineering challenges in Requirements Engineering phases and (iii) proposal to the use specific types of analytics to overcome identified Requirements Engineering challenges. Survey can be further researched collecting publications from different years, industries, main topics and use cases. Use of analytics in Requirements Engineering has a big potential, but nowadays use of analytics in Requirements Engineering is still limited, the issues addressed in this work are rarely surveyed, structured and organized, so the knowledge of the use of analytics in Requirements Engineering could be reused and utilized effectively. In previous research [16], first insights of use of analytics in Requirements Engi- neering were collected and this research is a step towards the effective utilization of different types of analytics in Requirements Engineering. The presented research has several limitations – more sources of analytics and Re- quirements Engineering challenges can be identified, analytics can be grouped by the usage and use of analytics in Requirements Engineering challenges can be explained using examples. Nevertheless, the contribution of this research in progress provides insights about possibilities of use of analytics in Requirements Engineering. The further research will include overcoming of the above listed limitations. References 1. Chaffey D. & Patron M. From web analytics to digital marketing optimization: In- creasing the commercial value of digital analytics. In Journal of Direct, Data and Dig- ital Marketing Practice. Vol. 14. Springer, 30-45 (2012). 2. Dankov Y., Birov D.: General Architectural Framework for Business Visual Analyt- ics. Business Modeling and Software Design, Springer, 280-288 (2016). 3. Dick J., Hull E. & Jackson J. Requirements Engineering: Fourth Edition. Springer, p.239 (2017). 4. Eradze M., Valjataga T. & Laanpere M.: Observing the Use of e – Textbooks in the Classroom: Towards “Offline” Leaning Analytics. In proceedings on International Conference on Web-Based Learning (ICWL 2014), pp. 254 – 263 (2014). 5. Gartner. Gartner Glossary: Diagnostic Analytics. Available at: https://www.gartner.com/en/information-technology/glossary/diagnostic-analytics. Accessed – 23.01.2020. 6. Ghozali R.P., et al.: Systematic Review on Decision-Making of Requirements Engi- neering from Agile Software Development. In the proceedings of the 4th International Conference on Computer Science and Computational Intelligence 2019 (ICCSCI), September 12-13, 274-281 (2019). 7. Hamilton R.H. & Sodeman W.A. The questions we ask: Opportunities and challenges for using big data analytics to strategically manage human capital resources. Business Horizons Vol. 63. January –February, 85-95 (2020). 79 8. Han L., et al. Who will use augmented reality? An integrated approach based on text analytics and field survey. European Journal of Operational Research. Vo. 281. 502- 516 (2020). 9. Kahan E., Genero M., Oliveros A. Challenges in Requirements Engineering: Could Design Thinking Help? In proceeding of the International Conference on the Quality of Information and Communications Technology (QUATIC 2019), 79 -86 (2019). 10. Karlsson L., et al.: Requirements Engineering challenges in market-drives software development – An interview study with practitioners. Information and Software Technology, Vol. 49, 2007, pp.588-604 (2007). 11. Mantymaki M., et al. How Do Small and Medium – Sized Game Companies Use An- alytics? An Attention- Bases View of Game Analytics. In Information Systems Fron- tiers, Springer (2019). 12. Marcelino – Jesus E., et al.: A Requirements Engineering Methodology for Techno- logical Innovations Assessment. In proceedings of the International Conference on Concurrent Engineering (CE 2014), p. 11 (2014). 13. McCarthy R.V. et al. Introduction to Predictive Analytics. In Applying Predictive Analytics. Springer, pp.1-25 (2019). 14. Mishra V., et al. Use of Visual Analytics and Durometer in Risk Reduction of Foot Problems in Diabetes. Lecture Notes in Mechanical Engineering. Springer, pp. 491- 498 (2016). 15. Ngo J., et al. Factor – based big data and predictive analytics capability assessment tool for the construction industry. Automation in Construction. Vol. 110 (2020). 16. Pincuka M., Kirikova M.: Types of Analytics in Requirements Engineering. In pro- ceedings of Business Informatics Research (BIR2018), pp. 25-32 (2018). 17. Pincuka M.: Master thesis: “Analytics in Requirements Engineering”, 2019, p.98. 18. Predictive Analytics Today. What is Text Analytics? Available at: https://www.predictiveanalyticstoday.com/text-analytics/. Accessed – 23.01.2020. 19. Reifferscheid K. & Zhang X. Enhance the Use of Medical Wearables Through Mean- ingful Data Analytics. In proceeding of International Conference on Digital Human Modeling and Application in Health, Safety, Ergonomics and Risk Management, Springer, pp. 281 -296 (2017). 20. Safwat A. & Senousy M.B.: Addressing Challenges of Ultra Large Scale System on Requirements Engineering. Procedia Computer Science, Vol. 65, pp.442-449 (2015). 21. Sawyer P., et al.: How the Web of Things Challenges Requirements Engineering. In proceedings of the International Conference on Web Engineering (ICWE 2012), pp. 170-175 (2012). 22. Schmid K.: Challenges and Solutions in Global Requirements Engineering – A Liter- ature Survey. In proceedings of International Conference on Software Quality (SWQD 2014),pp. 85-99 (2014). 23. Schon E-M., et al.: Key Challenges in Agile Requirements Engineering. In proceed- ings of International Conference on Agile Software Development (XP 2017), pp.37- 51 (2017). 24. Sun Z., Pambel F. & Wang F., Incorporating Big Data Analytics into Enterprise In- formation Systems. In proceedings of Information and Communication Technology – EurAsia Conference (ICT-EurAsia2015), pp. 300- 309 (2015). 80