Elicitation of Needs Using User Personas to Improve Software User Experience

— Tweets from users in the form of opinions about a product can be used as a company evaluation of the product. To obtain this evaluation, the method that can be used is sentiment analysis to divide opinions into positive and negative opinions. This study uses 1000 data from Twitter related to an internet service provider company where the data is divided into two classes, namely 692 positive classes and 308 negative classes. In the Tweet there are still many words that are not standard. Therefore, previously carried out the initial process or preprocessing to filter out non-standard words. Before doing the classification, the data needs to be divided into training data and test data with a ratio of 90:10, then processed using the Support Vector Machine and Naïve Bayes techniques to get the results of the classification of positive opinions and negative opinions. The level of accuracy in the classification using the Support Vector Machine is 84% and using Naïve Bayes is 82%.


INTRODUCTION
The development of technology is currently very fast and will continue to grow, it has a huge impact on various aspects that exist. Many industrial fields are finally competing to develop technology in these fields [1]. This is because it is very helpful to increase effectiveness and efficiency in work processes within a company or organization [2]. Thus, it can be concluded that today's information technology is very important in the success of an organization or agency management [3]. In order to create an effectiveness and efficiency, proper governance of information technology is needed [4]. Thus, "organizational and agency information technology is related to how top management of organizations and agencies gain confidence that the Information System Manager (Chief Information Officer) and IT organizations can provide returns or returns in the form of value for the organization or agency [5]." In other words, information technology can provide support to organizations/agencies in achieving the desired goals [6].
Information system analysis is very important to do, it is needed in order to produce a system that is easy, effective, efficient, and also appropriate for the users of the system. One model of information systems analysis by applying the concept of Human Computer Interaction (HCI) [7].
Techniques in software development are growing with the times, one of which is widely used, namely the Human Computer Interaction (HCI) approach. Producing a usable and safe system in the sense that the system can function properly is the goal of HCI, while also placing more emphasis on evaluation, design and implementation aspects. In several previous studies on HCI the HCI model can help analyze how users can interact with computers, so that they can describe behavioral models from various points of view [8].
The software development process is a phase that is often researched, including the excavation and requirements analysis phase. The purpose of this stage is to obtain specific requirements, both functional and non-functional. Several approaches have been carried out by several previous researchers. Requirements analysis is a process to get user requirements and constraints clearly and precisely [9], software requirements will be needed through intensive communication to get user needs [10], in order to eliminate ambiguity that often occurs due to the inability to explore these needs. or also because stakeholders (stakeholders) cannot properly express the desired needs.
Several approaches to exploring needs include group methods and individual methods, one of which is to group customer needs into 5 segments through adscore [11]. The group method is often used for several similar stakeholders and usually in large numbers, while the individual method focuses more on person to person within these stakeholders. Previous research conducted individual interviews from several experts with the aim of obtaining model requirements [12].
When developing an information system, the developer of the system must pay attention to the human-computer interaction factor. Human and Computer Interaction (HCI) is a scientific discipline that examines communication or interaction between users and systems. The main role of HCI is to produce ISSN 2085-4552 a system that is useful, safe, productive, effective, efficient and functional [13].
HCI focuses more on system design on the user or User Center Design (UCD), by paying attention to the users who will use the system, the system can be created according to and appropriate for the user. A system that is right for the user will provide comfort to the user when using the system so that the purpose of implementing the system will be achieved and will not fail [13] .
In this case study, the needs analysis was obtained from stakeholders related to or related to the practicum module uploading system through infotech which is within the academic scope of the University of Muhammadiyah Malang. In this study, it involved stakeholders from students majoring in informatics class 2019 who took part in practicum activities, lecturers who were in charge of courses who had practicums, and laboratory assistants who supervised the practicum. The current practicum module uploading system is considered less effective due to the sometimes late uploading of practicum modules so that students or practitioners lack time to study and complete the given practicum assignments.
In this study, we will focus on exploring existing needs through the human-computer interaction (HCI) approach, then collecting data on characteristics from stakeholders related to the existing practicum module uploading system. Persona taken later will be in the form of characteristics and objectives based on the results of interviews and observations [14]. Quantitative and qualitative data will later become the basis for conducting analysis and synthetics to be collected as persona designs [15]. This method will produce an approach in various aspects of psychology, objectivity, motivation and existing habits, based on demographics and social. Furthermore, for the characteristics of Application / service, Skills / Previous Experiences, and Goals can help identify possible application requirements. However, these characteristics are so general that they may present information that is not part of the application domain and may not result in the requirements. Attitudes and behaviors are important to describe in personas, they are needed to understand and identify possible requirements that are useful for application development [16].

II. RELATED RESEARCH
Therefore, it can be identified that some of the limitations in the techniques listed in the table are: • The technique focuses on understanding rather than identifying possible requirements • Characteristics that help identify behavior are not focused specifically in the application domain, as these characteristics are described in general So it is necessary to propose specifically the techniques used to guide the description of personas that focus on identifying possible requirements related to the application domain to be developed.

A. Data Crawling
Crawling is a method of collecting data from a source to be analyzed or processed. Crawling is the first stage that is usually used to analyze the sentiments of social media users towards a product or service. Crawling can also be interpreted as a method of quickly gathering large numbers of web pages into a local storage and indexing them based on specified keywords [8]. Sentiment analysis usually uses the Crawling method on Twitter social media by utilizing the features provided by Twitter, namely Application Programming Interface Systems (APIs) [9].

A. Activity 1: State Hypotheses
State Hypotheses are formulated to identify the possibility of persona. This hypothesis helps to determine who will be the subject of the interview. This is the first stage in getting acquainted with users [17]. In addition, at this stage also carried out stages to design a formulation in the interview.

B. Activity 2: Identify Behavioral Variables
Identify a list of behavioral variables to characterize possible users taken from the results of the interview synthesis [17]. Which interview subjects will be easier to map if the range of values has been defined before mapping [16].

C. Activity 3: Identify Significant Behavior Patterns
Identify specific subject groups that appear in the range leading to the use of archetypes for personas and modeling users. The purpose of this activity will be to get a result in the form of two products, namely, the percentage grouping table and the arrangement of significant behavioral patterns, the percentage grouping table containing the percentage of interview subjects that have been mapped to each behavioral variable value [17].

D. Activity 4: Map Interview Subjects to Behavioral
Variables Represents how many subjects are grouped with respect to significant behavioral variables, which is the initial model of system users [17]. This activity aims to detect the grouping of certain behavior patterns with the existing range of values [17], so that they can be made significant and become the basic pattern of the persona.

E. Activity 5: Synthesize Characteristics and Relevant
Goals This activity generates a basic persona document. In short, the description of the relevant characteristics and goals reflects the personality of the created persona. This document is an end-user model document [17].

F. Activity 6: Check for Redundancy and Completeness
Examine the redundancy, completeness, mapping, characteristics and goals of personas to find out if there are important gaps [17]. To do this, we check that the essential identification and aspects are fully defined in the created personas. Looking for important information from the previous activity stages.

G. Activity 7: Expand the Description of Attributes and Behaviors
Extends Attribute and Behavior Descriptions in a narrative manner. Narrative is a clearer way to convey or communicate people's attitudes, needs and problems.
Many aspects of the Persona Document are obscure, and useless for this task as Narrative [17]. The narrative can also be with additional data by interviewing stakeholders

H. Activity 8: Designate Persona Types
Determine the persona we want to create by determining the primary and secondary types [17]. After describing all the personas in a narrative, in the next activity we need to determine which persona is the main target of the system being built. We must define only one of all the personas whose needs and objectives can be met and represented in one interface without upsetting the other personas. To determine which persona will be the main target of the system, we must divide the persona into 2 types, namely primary and secondary personas. The primary persona is the persona that represents the main needs & objectives of the system development. While the secondary persona is a person who has additional needs that are not in the primary persona.

I. Activity 9: Build Use Cases
The activity that must be done after the persona that is the main target is clear is to create a use case. Use cases are made based on the persona in the basic document as well as in the narration as well as information from users obtained from previous activities [17]. Use cases are made in the form of diagrams like use case diagrams in general, which are made using annotations that show the relationship with primary and secondary personas. And also made a use case scenario or the flow of the prototype that we will make in the next stage.

J. Activity 10: Mockup
Implement and Evaluate Prototypes. We have used the use case developed in Activity 9 to implement the prototype.
Creating mockups designed by stakeholders [17]. Mock Ups are made based on use cases and also various needs that have been recorded in basic documents in previous activities. Then from the mock-up that is made, we evaluate the environment in which the system will be implemented, because there we will find potential users of the system being built.

A. Activity 1: State Hypothesis
The purpose of this activity is to identify the variables that can be created, to find out the differences between users based on their needs and behavior. Helping students to easily understand the material that has been given From the table above shows a list of hypotheses for each persona, from the results obtained, it can be identified and made personas synthesize between students, lab assistants and lecturers to assemble the necessary needs.

B. Activity 2: Identify Behavioral Variables
The results of the interviews will be given a range with values that have been defined before the mapping is made. From the table of behavioral variables above, it can be identified various characteristics of each persona. Which is included in the range of values that have been set in an infotech system access activity, which is in accordance with the existing problem.

C. Activity 3: Identify Significant Behavior Patterns
At this stage, each variable from each range of values will be grouped. From the results above, the percentage will be made to get a significant grouping of each behavior pattern of the existing respondents. To get the percentage, the results of the values obtained from each subject will be calculated in comparison with the total number of existing subjects. For the results of respondents' answers taken from students, in the first variable the higher percentage is very often, then for the second variable the higher percentage is often, and for the last variable the higher percentage is normal. For the results of respondents' answers taken from students, in the first variable the higher percentage is Infrequently, then for the second variable the higher percentage is Infrequently, then for the third variable the higher percentage is never, and for the last variable the higher percentage is.

ISSN 2085-4552
For the results of respondents' answers taken from students, in the first variable the higher percentage is often, then for the second variable the higher percentage is normal, then for the third variable the higher percentage is never, and for the last variable the higher percentage is Often.

D. Activity 4: Map Interview Subjects to Behavioral
Variables At this stage, fragment mapping will be carried out from the answers given from stakeholders based on the behavioral identification that has been made. Of the cases that have been identified, there are 5 respondents each from each stakeholder who have provided answers.

Fig. 2. Student Mapping Fragments
In the mapping fragment image above, the results of each variable choice for student respondents are obtained. For the first question, how many times have you used infotech, the answer is very often, because all UMM informatics students in each semester must take at least 1 course that has a practicum. So they need to access infotech for practical needs such as accessing modules to study and do assignments from the given practicum, this is because the modules that are distributed sometimes have not been uploaded so some students need to check whether the modules have been uploaded or not. There is also the use of the file attachment feature which is commonly used by students to collect practicum assignments. Furthermore, for respondents from lecturers. For the first question, namely how many times have used infotech, the answer is infrequently or only occasionally. Furthermore, for the features used, namely module access which is sometimes used to view modules so that they can be taught a little to students.
Then the attachment feature is never used by lecturers because they do not collect practicum assignments and upload modules. And the last one is the assessment feature where every lecturer never uses this feature because the practicum assessment is submitted to the lab assistant who will later pay the grade to the lecturer. The last one is the lab assistant respondent. For the first question, namely how many times have you used infotech, the answer is often, because the lab assistant is responsible for carrying out the practicum of each practitioner. Furthermore, the features used are module access which is often used to view the modules that will be used by the practitioner when doing practicum. Then for the use of the attachment feature, the lab assistant uses the attachment feature to upload the module, but sometimes only the CO of the lab assistant from each class uploads the module. And the last is the assessment feature, this feature is often used by every lab assistant when conducting practicums to assess practitioners. In the picture above, the results obtained from each choice of variables for student respondents. For the first question, how many times have you used infotech, the answer is very often, because all UMM informatics students in each semester must take at least 1 course that has a practicum. So they need to access infotech for practical needs such as accessing modules to study and do assignments from the given practicum, this is because the modules that are distributed sometimes have not been uploaded so some students need to check whether the modules have been uploaded or not. There is also the use of the file attachment feature which is commonly used by students to collect practicum assignments. Furthermore, for respondents from lecturers. For the first question, namely how many times have used infotech, the answer is infrequently or only occasionally. Furthermore, for the features used, namely module access which is sometimes used to view modules so that they can be taught a little to students. Then the attachment feature is never used by lecturers because they do not collect practicum assignments and upload modules. And the last one is the assessment feature where every lecturer never uses this feature because the practicum assessment is submitted to the lab assistant who will later pay the grade to the lecturer. From the four questions that have been given, it can be seen that there is one fragment that shows the significance of the choice of all respondents. For the existing fragments, the significant respondents chose the seldom choice and never chose the option. The last one is the lab assistant respondent. For the first question, namely how many times have you used infotech, the answer is often, because the lab assistant is responsible for carrying out the practicum of each practitioner. Furthermore, the features used are module access which is often used to view the modules that will be used by the practitioner when doing practicum. Then for the use of the attachment feature, the lab assistant uses the attachment feature to upload the module, but sometimes only the CO of the lab assistant from each class uploads the module. And the last is the assessment feature, this feature is often used by every lab assistant when conducting practicums to assess practitioners. From the four questions that have been given, it can be seen that there are two fragments that indicate the significance of the choices of all respondents. For the existing fragments, the respondents consistently chose the frequent choice, the usual choice and the never choice.

F. Activity 6: Check for Redundancy and Completeness
At this stage, a check will be carried out regarding the behavioral variables that have been made by adjusting the results of the behavioral variables that have been carried out with the behavioral variables from the new respondents. This is done to find out whether the needs analysis carried out is appropriate or still requires additional additions.    There is a document that describes details of the activities carried out for each use case. Specifications are described in the form of use case descriptor that provides a description more about use case, result of use case descriptors are described by Table VII.   TABLE VII. IDENTIFY SIGNIFICANT BEHAVIOR PATTERNS (LAB ASSISTANT)
Modul di publish sesuai waktu The final results at the research stage that have been carried out are made in the form of a mockup according to the evaluation carried out with the related persona. In Figure 12 is the start page where there is a button to upload the practicum module. Then in the 13th picture, after the button to upload the module is pressed, a menu for uploading the practicum module will appear and also a menu for setting the schedule for uploading the practicum module so that the practicum module will be uploaded according to a predetermined schedule.

V. CONCLUSIONS
In the research conducted, it was found problems related to the module uploading system through infotech, namely the upload of the practicum module which is sometimes late and also the absence of a notification feature when the practicum module has been uploaded. practice has been uploaded.
To overcome this, a solution that might help is needed is the addition of a scheduled upload feature so that later the practicum module uploader can be done automatically according to a predetermined schedule and also the addition of a notification feature when the practicum module has been uploaded.