Binus University Repository

Girsang, Abba Suganda

Reprint From 
 
Electronic ISBN: 978-1-5386-2182-0, Print ISBN: 978-1-5386-2180-6, USB ISBN: 978-1-5386-2181-
3 Print on Demand (PoD) ISBN: 978-1-5386-2183-7 
https://ieeexplore.ieee.org/abstract/document/8304103 
2017 International Conference on 
Sustainable Information Engineering and 
Technology (SIET) 
 
Sentiment Analysis of Customer Engagement on 
Social Media in Transport Online 
 
 
Melva Hermayanty Saragih; Abba Suganda Girsang,  
Computer Science Department, Binus Graduate Program,  Bina Nusantara University, Jakarta, Indonesia, 
 
 
 
DOI  
10.1109/SIET.2017.8304103 
 
 
 
 
  
 
 
 
 
 
 Conference Committees 
Honorary Chairs: 
• Wayan Firdaus Mahmudy, Universitas Brawijaya, ID 
• Nazri bin Mohd. Nawi, Universiti Tun Hussein Onn Malaysia 
(UTHM), MY 
• Pitoyo Hartono, Chukyo University, JP 
International Advisory Boards: 
• Toru Matsumoto, The University of Kitakyushu, JP 
• Shahreen binti Kasim, Universiti Tun Hussein Onn malaysia, MY 
• Mohd Farhan bin Md Fudzee, Universiti Tun Hussein Onn 
malaysia, MY 
• Hairulnizam bin Mahdin, Universiti Tun Hussein Onn Malaysia, 
MY 
• Ade Gafar Abdullah, Universitas Pendidikan Indonesia, ID 
• Tri Astoto Kurniawan, Universitas Brawijaya, ID 
• Son Kuswadi, Politeknik Elektronika Negeri Surabaya, ID 
• Wisnu Jatmiko, Universitas Indonesia, ID 
• Heru Nurwarsito, Universitas Brawijaya, ID 
• Yudho Giri Sucahyo, Universitas Indonesia, ID 
• Satriyo Dharmanto, IEEE Indonesian Section, ID 
• Syeilendra Pramuditya, Institut Teknologi Bandung, ID 
General Chairs: 
• Tibyani TIBYANI, Universitas Brawijaya, ID 
• Azizul Azhar Ramli, Universiti Tun Hussein Onn Malaysia (UTHM), MY 
• Donny Koerniawan, Institut Teknologi Bandung, ID 
Co-Chairs: 
• Gembong Edhi Setyawan, Universitas Brawijaya, ID  
Technical Program Commitee Chairs: 
• Achmad Basuki, Universitas Brawijaya, ID 
• Dahnial Syauqy, Universitas Brawijaya, ID 
• Sigit Adinugroho, Universitas Brawijaya, ID 
Publication Chairs: 
• Imam Cholissodin, Universitas Brawijaya, ID 
Financial Chairs: 
• Sri Suyati Luhur Utami, Universitas Brawijaya, ID 
• Heru Dintanufi, Universitas Brawijaya, ID 
Local Organizing Committee: 
• Yuita Arum Sari, ID 
• Denny Sagita Rusdianto, ID 
• Adhitya Bhawiyuga, ID 
• Mochammad Hannats Hanafi, ID 
• Rakhmadhany Primananda, ID 
• Hurriyatul Fitriyah, ID 
• Rizal Maulana, ID 
• Randy Cahya Wihandika, ID 
• Budi Darma Setiawan, ID 
• Muhammad Tanzil Furqon, ID 
• Widhi Yahya, ID 
 
  
  


English

Commitee "International Conference on Sustainable Information Engineering and Technology, SIET 2017"

Sasmoko, Sasmoko

Lembar Peer Review "Building a popular mobile application by utilizing user feedback"

Wahyono, Teguh

Spits Warnars, Harco Leslie Hendric

Wijaya, Billy Sentosa

Fahri, Ahmad

Matsuo, Tokuro

Building a popular mobile application by utilizing user feedback

Rohayani, Hetty

Lumban Gaol, Ford

Soewito, M.Sc., Ph.D, Benfano

2017 International Conference on Applied Computer and Communication Technologies (ComCom), Jakarta, Indonesia
Estimated Measurement Quality Software
On Structural Model Academic System
With Function Point Analysis
Hetty Rohayani
Doctor of Computer
Science Program
Bina Nusantara University
n. Kebon Jeruk No27,
Jakarta.
Indonesia
hetty_mna@yahoo.com
Ford Lumban Gaol
Doctor of Computer
Science
Bina Nusantara University
n. Kebon Jeruk No27,
Jakarta.
Indonesia
fgaol@binus.edu
Benfano Soewito
Binus Graduate Program
Bina Nusantara University
n. Kebon Jeruk No27,
Jakarta.
Indonesia
bsoewito@binus.edu
Harco Leslie Hendrie
Spits Warnars
Doctor of Computer
Science
Bina Nusantara University
JI. Kebon Jeruk No27,
Jakarta.
Indonesia
shendric@binus.edu
Abstract- In the software development indispensable is the
suitability and accuracy in determining the size or value of the
software to fit the operation to be performed. A wide variety of
calculation methods have been widely used to estimate the size of
the software, one of which is by using Function Point Analysis
(FPA). Volume calculation software based on a scale of
complexity. Since the point of measurement is highly subjective,
in order to maintain consistency and validity of the results, the
method should be run by an experienced professional. This
method is then applied by the authors to measure the complexity
of academic information system STIKOM Dinamika Bangsa
Jambi using structured modeling approach. Measurements were
performed in this study consisted of depictions information
system is built into the structure. Which is then analyzed by
counting models Crude Function Points (CRP), the relative
complexity of Adjustment Factor (RCAF), and then calculate the
point function. From the results of calculations using the FPA to
software quality measurement academic system STIKOM
Dinamika Bangsa Jambi obtained value FP 166.32 is good.
Function point value produced will be used by developers in
determining the price and the cost of software systems to be built
or developed.
Keywords- Software Quality Measurement; Structural Model;
Function Point Analysis; System
I. INTRODUCTION
The success of a software project is determined by
many factors that are interconnected in a project. A project
will be declared a success if all the existing requirements can
be met, including the cost is not excessive (overflow), the
schedule not exceeding the time limits specified. This can be
done either by estimating the size of the volume of software
systems that will provide accuracy the complexity and value
of the product price on the software project being built. So that
developers can plan resources, project costs, and the duration
required precisely to build up a good software [1],[2].
Software development can be defined as the process
of creating a new software to replace the old software as a
whole or improve existing software. To be faster and more
accurate in describing and developing software solutions, as
well as the results easily developed and maintained, then the
software development requires a specific methodology.
Software measurement methodology is a process of organizing
a collection of methods and notation conventions that have
been defined to measuring the software [3]. The use of a
methodology in accordance with the problems to be solved
and meet the needs of users will produce a product
engineering quality and well-maintained, and can avoid the
problems that often occur as scheduling and cost estimation,
software that does not comply with the wishes of users and so
on.
For the developer, measuring the volume of useful
software for resource planning, cost and duration necessary to
build a software [3]. In addition, developers can also evaluate
the quality of a product by comparing the volume of the
system with the number of errors (error-count) done in
software. Meanwhile, from a business perspective, the volume
of software can be the basis for determining the value of the
price of the software in question.
Tn the past many measures the volume of a software
using a simple method that LOC (Lines Of Code) or KLOC
(1000 Lines Of Code) [4],[5], which is a software engineering
major measurement by counting lines of existing code. This
technique has the properties he lacked, namely:
1. Relative to the language / programming tools and
programmers coding style. LOC is highly dependent on
the characteristics of programming tools used and
programmers coding style. For example in the BASIC
language code as follow:
a=a+l
978-1-5090-4048-3/17/$31.00 ©2017 IEEE
2017 International Conference on Applied Computer and Communication Technologies (ComCom), Jakarta, Indonesia
only requires one line of code. Meanwhile, to get
the same results in the PASCAL language the code
is converted as follows:
program
x; var
a: integer;
begin
a:= a+ 1;
end.
which requires six lines of code. Also note the example of the
difference of two script coding style following courses that
lead to differences in LOC
Code 1 (2 line) Code 2 (4 line):
a:=a+l; b:=5; a:=a+l;
if a=2 then a:=I; b:=5;
if a = 2 then
a:= 1;
6. LOC can not be determined prior to completing the
project development stages of implementation (coding).
Therefore, the LOC can not be used to plan development
process and also can be used to estimate the price of the
product. The completion of the implementation phase is
a phase that is too late to prepare a resource estimate.
The case of shortage then the desire to obtain a software
volume measurement techniques that are not just based on the
number of lines of programming code, but more towards
something that can be measured early in the software
development life cycle so that later came out the idea of
method Function Point. In this study, the authors will discuss
the use of the methods Function Analysis Point (FPA) to
measure the estimated volume of the software system, the
system of academic education Stikom Dinamika Bangsa Jambi
will be compared to the estimated model using Entity Relation
Diagram (ERD) based object model with the data flow
Diagrams (DFD) based structured model.
2. LITERATURE REVIEW
a. Function Point Analysis (FPA)
FPA method is a part of The FSM (Functional Size
Measurement) method was flrst introduced by Albrecht in
1979 as a method for measuring the amount of complexity and
functionality in a software project [7],[8],[9],[10]. In the FPA
procedure there are a variety of transactions, comprising the
incoming and outgoing data to be processed on the system.
Each transaction in the FPA will be mapped to the following
models.
In the picture above can be seen, The Function Point Model
consists of [6] :
1. External Input (EI)
Functions that move data into the application without
presenting data manipulation.
2. External Output (EO)
Functions that move data to user and presents some data
manipulation.
3. External Inquiries (EQ)
Functions that move data to user without presenting data
manipulation.
4. Internal Logical Files (ILF)
The logic in the form of flxed data managed by the
application through the use of External Input (EI)
5. External Interface Files (ElF)
The logic in the form offlxed data used by the application
but did not run in it.
FUlletlon Polnl Mood
~".1
-
Fig 1: Function Point Method [11]
In the computation phase, each transaction is sorted by the
amount of data that use. Logical transaction or fIle sorted
based on the entities (called RET or Referenced Entity Types)
and attributes (called DET or Data Entity Types). Functional
transaction sorted based on the attribute numbers (DET),
which moved out of the line and the numbering of logical
transaction references. Then the whole categorized into 'low',
'average', or 'high' which each category given value as the
value of Function Point (FP)[12],[13].
The stages are there in determining the function point is
[14],[15],[16].
Step 1: Calculating The CFP (Crude Function Points)
The number of functional components of the system were
flrst identified and followed to evaluated the complexity of
quantization weight of each component. Weighting was the
summed and become the number of CFP. CFP calculation
involves five types of software system components
following:
1. The number of input applications
2. The number of output applications
3. The number of online query applications.
4. This application related to query against the data
stored
5. The number of logical files or tables which involved
6. The number of external interfaces
An output or input interface that can be connected to the
computer through data communications, CDs, floppy
2017 International Conference on Applied Computer and Communication Technologies (ComCom), Jakarta, Indonesia
disks, etc. Then given a weighting factor to each of the
above components based on its complexity. The table
below is an example of the weighting blank:
Table 1. Blank of CFP Calculation 191
Software Complexity Level Total
System Simple Average Complex CFP
Component C W P C W P C W P
Inputs - 3 - - 4 - - 6 - -
Outputs - 4 - - 5 - - 7 - -
Online Queries - 3 - - 4 - - 6 - -
Logical Files - 7 - - 10 - - 15 - -
External Interfaces - 5 - - 7 - - 10 - -
Total CFP -
Step 2: Calculating the complexity of transcription factors of
RCAF (Relative Complexity Adjustment Factor) for the
project. RCAF is to calculate the complexity assessment of
software system from several characteristics of subject. Rating
scale from 0 to 5 is given to each subject that most affect the
development effort required. Example of RCAF assessment
form can be seen as follows:
Table 2. RCAF Assessment Form [9]
No Subject Value
1 he level of recovery reliabilty complexcity 1 2 3 4 5
2 he level of data communication complexcity 1 2 3 4 5
3 he level of distributed processing complexity 1 2 3 4 5
4 Level of the need for performance complexity 1 2 3 4 5
5 he level of operating environment demand 1 2 3 4 5
6 he level of developer knowledge needs 1 2 3 4 5
7 he level of updating the master file complexity 1 2 3 4 5
8 he level of rotallection complexity 1 2 3 4 5
he level of input, output, online query and file
9 ~pplication complexity 1 2 3 4 5
10 he level of data processing com plexity 1 2 3 4 5
11 he improbability level of reuse code 1 2 3 4 5
12 he level of custumer organization variation 1 2 3 4 5
13 he expert of possible changes 1 2 3 4 5
14 Level of the ease of use demand 1 2 3 4 5
Total RCAF
Step 3: Calculating Function Points by the formula [9].
FP = CFP x (0.65 + 0.01 x RCAF) (1)
Values for the function point software system is then
calculated based on the results of phase 1 and 2 are put into
the formula.
b. Structural Model
In DFD, there are some important components which
is component as terminators in a box of a source or
destination, the components in the form of a circle of as a
description of the activities that will be/are being
implemented conslstmg of input and output, data store
component in the form of two horizontal lines are lined up as
an image of computerized storage, and data flow components
depicted in the form of an arrow as data transfer from one
section to another system [17].
In ERD, the modeling made up of several
components, namely the entities, attributes, relationships,
cardinality ratio and constraint participant. An entity is an
object that will store, produce information. Attributes are the
characteristics of the entity or relationship that provides a
detailed explanation of the entity or relationship. Relationship
is the relationship between one or more entities. The
cardinality ratio is a limit to the number of connection one
entity to another entity. While the constraint participant clarify
whether the existence of an entity depends on its relationship
with other entities [18].
3. RESULT AND DISCUSSION
In this research, measurement of software quality
system of academic information STIKOM Dinamika Bangsa
Jambi, where the system will provide services such as filling
Study Plan Card or KRS online, structured approach in a paper
done by using the model of data flow diagram (DFD) and
Entity Relationship diagram (ERD). DFD is a diagram to
describe the system as a network function processes are
connected to each other by data flow. While ERD is a model
of the logical structure of the database lecture schedule, and
check the value of the program.
Fig 2. Data Flow Diagram of Academic System STIKOM
Step 1 : Calculate Function Point Crude
Based on the image below will be shown an example
of calculating the value of function point software system for
the proposed academic information STIKOM Dinamika
Bnagsa Jambi. From the image data flow diagram in Figure 2,
it can be determined the values needed to calculate the CFP,
which is the amount of each component, as follows:
2017 International Conference on Applied Computer and Communication Technologies (ComCom), Jakarta, Indonesia
• The number of input aplication: 4
• The number of output apiication: 3
• The number of online query: 9
• The number of logic file: 6
• The number of external interface: 0
After that, re-evaluated to sort each component based group
degree of complexity (simple, medium or complex) by a
professional to get the value of the CFP. In this experiment,
the results of these evaluations are presented in Table 3
which after calculation resulted in a total value ofCFP = 154.
b. Class Diagram
Fig 3. Entitas Relation Diagram of Academic System
STIKOM
Table 3. CFP Calculation of Academic System STIKOM
based on Structural Model
Software Complexity Level Total
System Simple Average Complex CFP
Component C W P C W P C W P
Inputs - 3 - I 4 4 3 6 18 22
Outputs - 4 - - 5 - 3 7 21 21
Online Queries 3 3 9 - 4 - 6 6 36 36
Logical Files - 7 - 3 10 30 3 15 45 75
External 5 7 10Interfaces - - - - - - -
Total CFP 154
Step 2: Calculate the relative complexity adjustment
factor(RCAF).
Table 4. RCAF Calculation of Academic System STIKOM
Dinamika Bangsa
No Subject Value
1 The level of recovery reliabilty complexcity 5
2 The level of data communication complexcity 4
3 The level of distributed processing complexity 3
4 Level of the need for performance complexity 4
5 The level of operating environment demand 3
6 The level of developer knowledge needs 3
7 The level of updating the master file complexity 2
8 The level of rotallection complexity 3
The level of input, output, online query and file
9 application complexity 2
10 The level of data processing complexity 3
11 The improbability level of reuse code 2
12 The level of custumer organization variation 3
13 The expert of possible changes 3
14 Level of the ease of use demand 3
Total RCAF 43
Then the function point of academic system
STIKOM Dinamika Bangsa Jambi based on structured
modeling can be calculated:
FP = CFP x (0.65 + 0.01 x RCAF)
= 154 x (0.65 + 0.01 x 43)
= 166.32
CFP calculation results can be used as the basis for
calculating large projects such as the time required for project
development, project costs, the level of resources required and
so on.
4. CONCLUSION
From the description and result of the experiments carried out
a number of conclusions:
1. Method of function point cab be used as an alternative to
calculate the volume of software based on its complexity.
2. The function point method has several advantages compared
LOC namely:
a. Does not depend on the programming language used
by programmers.
b. It does not depend on the style of programming a
programmer.
c. Can be determined early in the software development.
3. The use of function point method has the disadvantages of
requiring the intervention of an experienced professional for
calculation is done in a very subjective analysis.
4. Since the calculations are done a lot based on the
description of the data processing, methods of function
point should be supported by additional data to strengthen
the approximate volume of a software system that will be
generated.
The next step is to evaluate the complexity characteristics of
STIKOM Dinamika Bangsa Jambi and calculation of the
RCAF. RCAF calculation results in this study are described in
Table 4, there is shown a summation to the 14 th element /
factor adjustment relative complexity of generating value 43.
[1 ]
REFERENCES
J. G. Borade, "International Journal of Advanced
Research in Software Project Effort and Cost
Estimation Techniques," vol. 3, no. 8, pp. 730-739,
2017 International Conference on Applied Computer and Communication Technologies (ComCom), Jakarta, Indonesia
2013.
[2] F. L. Gaol, B. Soewito, H. Leslie, and H. Spits,
"Object Oriented Metrics to measure the quality of
software study case Request online System," 2016.
[3] M. Lee and T. Chang, "Software Measurement and
Software Metrics in Software Quality," vol. 7, no. 4,
pp. 15-34,2013.
[4] M. Sadiq, M. Asim, J. Ahmed, V. Kumar, and S.
Khan, "Prediction of Software Project Effort
Estimation: A Case Study," vol. 1, no. 1, pp. 37--43,
2011.
[5] K. P. Srinivasan, "Unique Fundamentals of Software
Measuremant And Software Metrics," vol. 7, no. 4,
pp. 29--43, 2015.
[6] D. Pratiwi and D. Pratiwi, "Implementation of
Function Point Analysis in Measuring The Volume
Estimation of Software System in Object Oriented ...
Implementation of Function Point Analysis in
Measuring the Volume Estimation of Software System
in Object Oriented and Structural Model of Academic
System," no. August 2016,2013.
[7] C. Gencel, M. East, A. Prof, and M. East, "Conceptual
Differences Among Functional Size Measurement
Methods."
[8] N. Choursiya and R. Yadav, "An Enhanced Function
Point Analysis ( FPA ) Method for Software Size
Estimation," vol. 6, no. 3, pp. 2797-2799, 2015.
[9] A. R. Irawati and K. Mustofa, "Measuring Software
Functionality Using Function Point Method Based On
Design Documentation," vol. 9, no. 3, pp. 124-130,
2012.
[10] G. Xunmei, S. Guoxin, and Z. Hong, "The
Comparison between FPA and COSMIC-FFP," pp.
109-117.
[11] T. Fetcke, A. Abran, T. Nguyen, and M. Q. C. Hc,
"Mapping the OO-Jacobson Approach into Function
Point Analysis."
[12] M. Heri and A. Zivkovi, "A Method for Calculating
Acknowledged Project Effort Using a Quality Index,"
vol. 31, pp. 431--436, 2007.
[13] T. Arnuphaptrairong, "The Development and
Achievements of Software Size Measurement," vol. I,
2012.
[14] W. Wikusna, "InSearch, Universitas Informatika dan
Bisnis Indonesia - 1 of9.," pp. 1-7.
[15] K. Kusrini, "Pengukuran Volume Software
berdasarkan Kompleksitasnya dengan Metode
Function Point," no. September, 2015.
[16] R. K. Hapsari and M. J. Husen, "Estimasi Kualitas
Perangkat Lunak Berdasarkan Pengukuran
Kompleksitas Menggunakan Metrik Function
Oriented," pp. 425--434, 2015.
[17] A. Elbakush, "Functional Modeling with Data Flow
Diagrams."
[18] H. K. Al-masree, "Extracting Entity Relationship
Diagram ( ERD ) From Relational Database Schema,"
vol. 8,no.3,pp. 15-26,2015.


Estimated measurement quality software on structural model academic system with function point analysis

Sefina Samosir, Ridha

Abdurachman, Edi

Measurement Metric Proposed For Big Data Analytics System

Trisetyarso, Agung

Prasetya Kristiadi, Dedy

Understanding of data mining in computer science learning from PILKADA DKI Jakarta 2017

Sianipar, Nesti Fronika

Reviewer Form Effect of gamma irradiation on protein profile of rodent tuber (typhonium flagelliforme lodd.) In vitro mutant based on 1D and 2D page analyses

Mulyadi, Martin Surya

Form Peer Review "Analysis of Corporate Income Tax Rate Changes and Earnings Management"

Form Peer Review "Influence of corporate governance and profitabilityn to corporate CSR disclosure"

RAPINTA MANALU, SONYA

Saleh Abbas, Bahtiar

Gaol, Ford L.

Lukas, Lukas

Trawinski, Bogdan

Form Reviewer "An expert system to assist with early detection of schizophrenia"

Binus University

Not a member yet