- Email: [email protected]
Mobile Smart Travelling Application For Indonesia Tourism
Available online at www.sciencedirect.com Available online at www.sciencedirect.com
Available online at www.sciencedirect.com
ScienceDirect
Procedia Computer Science 00 (2017) 000–000 Procedia Computer Science (2017) 000–000 Procedia Computer Science 11600 (2017) 556–563
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
2nd International Conference on Computer Science and Computational Intelligence 2017, 2nd International Conference on Computer Science andBali, Computational ICCSCI 2017, 13-14 October 2017, Indonesia Intelligence 2017, ICCSCI 2017, 13-14 October 2017, Bali, Indonesia
Mobile Mobile Smart Smart Travelling Travelling Application Application For For Indonesia Indonesia Tourism Tourism Meiliana, Devita Irmanti, Mochamad Rizal Hidayat, Nur Viani Amalina, Dewi Suryani∗∗ Meiliana, Devita Irmanti, Mochamad Rizal Hidayat, Nur Viani Amalina, Dewi Suryani
Computer Science Department, School of Computer Science, Bina Nusantara University, Jl. K. H. Syahdan No. 9, DKI Jakarta, 11480, Indonesia Computer Science Department, School of Computer Science, Bina Nusantara University, Jl. K. H. Syahdan No. 9, DKI Jakarta, 11480, Indonesia
Abstract Abstract Indonesia is a country that has seventeen thousand islands and located in a very strategic location. That reason makes Indonesia Indonesia a country that has seventeen thousand and located a very location. That makes of Indonesia becoming is a place that is interesting to be visited byislands many tourists eitherinlocal or strategic international. Every yearreason the number tourists becoming a place that is interesting to be visited by many tourists either local or international. Every year the number of tourists 1,2 The who visited Indonesia has greatly increased 1,2 and transformed the tourism to be an important sector for Indonesia economical. and transformed the tourism to be an important sector for Indonesia economical. The who visited Indonesia has greatly increased rapid development of tourism sector in Indonesia must be equilibrated by the evolution of technology and the supported facilities. rapid tourism sector in Indonesia be important equilibrated by the technology the activities. supportedBy facilities. Now, development the usage of aofsmartphone technology is onemust of the things andevolution a part ofof modern peopleand daily taking Now, the usage of of aa smartphone, smartphone technology one of thetoimportant things and aapplication part of modern activities. By taking the advantages this study isisconducted develop an android that ispeople nameddaily Smart Travelling. Smart the advantages of a smartphone, this study is conducted to develop an android application that is named Smart Travelling. Smart travelling has many features that can facilitate every tourist who visited Indonesia, such as recognizes any tourist attractions, shows travelling features can facilitate everythe tourist who visited Indonesia, such asfor recognizes any tourist shows any eventshas thatmany nearby visitedthat attractions, displays nearest police station and hospital an emergency case, attractions, and saves the hisany that nearby objects. visited attractions, displays nearest police station and hospital an emergency case, and saves theusing history events of the recognized The main idea of thethe tourist attractions recognition here is for to implement an image recognition tory of the recognized objects. The main idea of the tourist attractions recognition here is to implement an image recognition using 3 landmark detection feature from Google Cloud Vision Application Program Interface (API) 3 technology which helps the tourist landmark detection and feature from Cloudregarding Vision Application Interface technology helpsapplication the tourist to easily remember access anyGoogle information the visited Program attractions. In this (API) study, the result is awhich completed to easily remember and access any information regarding the visited attractions. In this study, the result is a completed application which is evaluated using questionnaires to 35 randomly selected participants. Based on the evaluation, the smart travelling achieves which evaluated questionnaires to 35agree randomly selected evaluation, smart travelling achieves positiveis inputs andusing mostly the participants that the imageparticipants. recognition Based featureonis the really helping the them. Furthermore, we also positive inputs and mostly the participants agree that the image recognition feature is really helping them. Furthermore, we also obtained a significant result for the scan landmark testing, which is 86% for the accuracy. obtained a significant result for the scan landmark testing, which is 86% for the accuracy. c 2017 2017 The The Authors. Authors. Published Published by by Elsevier Elsevier B.V. © B.V. c 2017 The Authors. Published by Elsevier B.V. Peer-review Computer Science Science and and ComPeer-review under under responsibility responsibility of of the scientific committee of the 2nd International Conference on Computer Peer-review under responsibility putational Intelligence 2017.2017.of the scientific committee of the 2nd International Conference on Computer Science and ComComputational Intelligence putational Intelligence 2017. Keywords: Image Recognition; Android application; Google Cloud Vision; Tourism; Keywords: Image Recognition; Android application; Google Cloud Vision; Tourism;
1. Introduction 1. Introduction Nowadays, the outgrowth of technologies and globalization era demand ease and automation in various aspects Nowadays, outgrowth of technologies and demand economic ease and automation in on various aspects of human life, the including tourism. In Indonesia, theglobalization tourism is anera important sector. Based the ministry of human life, including 1 tourism. In Indonesia, the tourism is an important economic sector. Based on the ministry of tourism data in 2016 1 , the number of local tourists in 2016 has been growing exponentially with over 6.6 million of tourism datahad in 2016 , the 1.97% numbercomparing of local tourists 2016 has beenFurthermore, growing exponentially over that 6.6 million people, which increased to the inprevious years. the foreignwith tourists visited people, which had increased 1.97% comparing to the previous years. Furthermore, 2 the foreign tourists that visited Indonesia had also increased 10.29% or approximately 10.4 million people in 2015 2 . Indonesia had also increased 10.29% or approximately 10.4 million people in 2015 . ∗ ∗
Corresponding author. Tel.: +62-21-534-5830 ext 2188; fax: +62-21-530-0244 Corresponding Tel.: +62-21-534-5830 ext 2188; fax: +62-21-530-0244 E-mail address:author. [email protected] E-mail address: [email protected]
c 2017 The Authors. Published by Elsevier B.V. 1877-0509 cunder 1877-0509 2017 The Authors. of Published by Elsevier B.V. the 2nd International Conference on Computer Science and Computational IntelliPeer-review the scientific committee 1877-0509 © 2017responsibility The Authors. Published Elsevier of B.V. Peer-review responsibility of the scientificbycommittee of the 2nd International Conference on Computer Science and Computational Intelligence 2017. under Peer-review under responsibility of the scientific committee of the 2nd International Conference on Computer Science and gence 2017. Computational Intelligence 2017. 10.1016/j.procs.2017.10.059
2
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
557
The rapid development of tourism sector in Indonesia must be equilibrated by the evolution of technology and the supported facilities. Now, the usage of a smartphone technology is one of the important things and a part of modern people daily activities. The utilities of the smartphone are not only in telecommunication but also in various fields, such as looking for information, entertainment, work supporter, data storage, photography, guidance, etc. By taking the advantages of a smartphone, this study is conducted to develop a mobile application. Currently, there are many kinds of platforms installed in a smartphone and the most popular platforms used are Android, iOS, and Windows Phone. Based on the number of smartphone users in 6 ASEAN countries 4 , most of the users used smartphones Android-based. Especially in Indonesia, at 2015, the Android users had reached 41 million users compared to 2.8 million iOS users. Therefore, this application is built in the Android smartphone. Indonesia is a country that is located mainly in South-east Asia and consists of seventeen thousand islands. Its strategic location makes Indonesia has many interesting sights to be visited by tourists both local and foreign. However, there is still not many applications that specific for all tourist attractions in Indonesia. In this paper, we report a tourism mobile application for Indonesia, which is named ”Smart Travelling” and is built only for the Android platform. The smart travelling has several functionalities, such as recognizes any tourist attractions, shows any events that nearby visited attractions, displays the nearest police station and hospital for the emergency case, and saves the history of the recognized objects. Furthermore, this application has a special feature that is not available in other similar applications. The feature is an image recognition that can be used to automatically recognize an object and give any information about it. The main purpose of this application is to help tourists to recognize any tourist attractions in Indonesia and get all related information. The remaining of this paper is organized as follows. In Section 2, we describe the relation of other studies for tourism purpose to ours. Next, all features in the proposed application and its methodology will be discussed in Section 3. Section 4 explains how we evaluate the application and its results. Finally, Section 5 concludes this paper and discusses the future development. 2. Related Works Tourism is an essential economic sector in a country. By focusing on tourism, many advantages can be achieved, such as an improvement in various sectors especially provision of employment, income, production, etc. Due to the necessity of tourism, several applications are built in order to endorse its process 5,6,7,8 . The work of Tobing 5 is one of the studies that conducts an android application for tourism in Samosir Regency, North Sumatera-Indonesia. It mainly focuses on the implementation of Google Map Application Program Interface (API) and integrated social media. Similar to Tobing 5 , Chao 6 developed the android application by integrating global positioning system (GPS). However, Chao 6 concentrated on Taiwan tourism and folk-cultural education instead of Indonesia tourism. Furthermore, Tian et al. 7 proposed an augmented reality technology in developing tourism application that can be used on any mobile platforms. In contrast to the other approaches, Bandung 8 implements digital signage system and Raspberry Pi 2 Model B as its client for supporting e-tourism in Indonesia. This study proposed an application for presenting Indonesia tourism with special features. Like Tobing 5 and Chao 6 , this application runs on the android platform. However, we focused on tourism area for entire Indonesia, which is a generalization of Chao 6 ’s approach. Unlike Tian et al. 7 , we applied the image recognition technique instead of augmented reality technology. Compared to the existing similar applications such as Pesona Indonesia and Wisata Lokal Travel Indonesia, our application has the combinations of all their features and offered a special feature which has not implemented in the others. The special feature here means the image recognition that can automatically recognize any tourist attractions which had taken by the tourist. Furthermore, our application also has a comprehensive tutorial which definitely helps the tourist to access and use our application in the proper way, and it is not limited in several Indonesia areas only. 3. System Overview As mentioned in the previous sections, the purpose of this study is to construct a mobile application for facilitating any tourists who are visiting Indonesia. The application is namely Smart Travelling and developed in Android platform. It has several main features that described in Fig. 1 and will be detailed as follows.
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
558
3
Fig. 1: Main menu of Smart Travelling contains scan landmark, event, emergency, history, tutorial, and profile features
3.1. Scan Landmark This feature has implemented the image recognition technology using Google Cloud Vision Application Program Interface (API) 3 . Google Cloud Vision API allows the developer to extract data from an image by encapsulating the machine learning model using REST API. It can classify any images into categories, detect individual objects, and read any printed words of an image. The Google Cloud Vision API can analyze either the uploaded images or the integrated images of the gallery in the Google Cloud Storage. There are many functionalities that are provided by the Google Cloud Vision API, however, this menu only applied the landmark detection feature, which identifies a natural or artificial object in an image. This feature not only focuses on recognizing the tourist attractions but also provides the additional information about it by automatically using Google search when the user requested as described in Fig. 2.
Fig. 2: (First) Scan landmark menu when the user selected it. (Second) Sub menu of scan landmark for performing image recognition using the camera. (Third) Sub menu of scan landmark for performing image recognition in the gallery. (Fourth) Scan landmark menu redirects to Google search engine when the user has chosen to search information regarding the recognized object
The scan landmark feature is the main focus of this study which takes the advantages of the landmark detection of Google Cloud Vision API. In this landmark detection, the first thing is getting the input images either via camera or by the gallery. Afterwards, the application sends a request to the Google Vision API for detecting the object. When the request is successful, the Google Vision API server returns the identifying information such as an entity ID, the landmark’s name, the landmark’s location, and the rectangle for showing the recognized object in the input image. However, the information that shown for the user is only the landmark’s name as shown in Fig. 9.
4
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
559
3.2. Event This feature provides any information about all ongoing or upcoming events in Indonesia which are gotten from the database that implemented using MySQL. Events data in the database were inputted by admin using a web-based application as the user interface and SQL code to manipulate the data. After the data is inputted, all events are viewed in Event Menu at the mobile version application as shown in Fig. 3. The user can view the detail location of the desired event based on Google Map API coordinate which is stored in the database too. There are also searching option to look for the event based on name or city which the application retrieve it from the user input and search the database using SQL command.
Fig. 3: Event menu while searching an event
3.3. Emergency This feature provides the nearest public services that users need immediately. The program reads the Android GPS current location and then search the nearest location of hospital or police station in radius 5 km using Google Maps API. It is depicted in Fig. 4. After the nearest hospital or police station is found, Google Maps API will return all coordinates of desired public service place and the fastest route to the position which will be displayed by the application.
Fig. 4: (Left) View of the emergency menu when the user selected it. (Center) The map displays all hospital’s coordinates after the user selected hospital sub menu. (Right) Map shows all police station’s coordinates when police station menu is selected
560
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
5
3.4. History The application implements a history feature which is used to view any landmark that has been scanned using the scan landmark menu. Everytime the user scans landmark, the landmark information will be stored into MySQL database. The data then will be displayed by the application using SQL code into History menu which already sorted by the latest scanned date. Fig. 5 displays the view of history menu when this menu is chosen.
Fig. 5: View of history menu when the user selected the menu
In addition, this application also provides tutorial and profile features. The tutorial feature displays all step procedures of each menu in the proposed application user interface in order to simplify the utilization of the application. The tutorial will provide all information for the user to use the application and described in Fig. 6.
Fig. 6: Tutorial menu for scan landmark, event, emergency, and history
Moreover, users can manage their own profile in the application through the profile menu, which is depicted in Fig. 7. The detail information of the user can easily be updated, such as name, email address, and phone number. Their password, which is used to login to the application, can also be changed. The application will update the new data of users to MySQL database using SQL command once the user confirms the changes. 4. Evaluation In order to evaluate the usage of the application and the user satisfactory of the smart travelling, this study has conducted white box and questionnaire evaluations after completing the application development process. White box testing is an evaluation method that depends on the internal structure of the application and executed during actual system operation 9 . Here, the white box evaluation will be done not only for front-end system but also for the back-end
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
6
561
Fig. 7: (Left) Profile menu in Smart Travelling. (Center) Update profile sub menu. (Right) Change password sub menu
component that had implemented in the web-based application. The results of these white box evaluations are depicted in Table 1 and Table 2. Table 1: The results of white box testing on the front-end of smart travelling application
Menu Register Login Scan Landmark Emergency History Tutorial Event Profile Logout
Function Description Register Login Scan landmark from camera and gallery Get tourist attraction name Search additional information of the attraction in Google Add the recognized object data into history menu Perform the emergency navigation Show the nearest location for emergency View the history data of recognized tourist objects Search additional information about the object in Google View information how to use the application View all ongoing and upcoming events information in Indonesia Search event based on the event name or the visited city View and update user data Change password Logout
Result OK OK OK OK OK OK OK OK OK OK OK OK OK OK OK OK
For the questionnaire evaluation, we provide 10 questions with the total number of participants is 35 people, which is randomly selected. Before filling the questionnaire, the participants need to install the application on their smartphones. At first, the user needs to download the smart travelling application in the Play Store. It can be done in two ways, i.e., search by keyword ”Smart Travelling” or click this URL ”https://play.google.com/store/apps/details?id=com.skripsi.smarttravelling”. Afterwards, install it and wait until the process is finished, then the application can be tested. Fig. 8 shows the questionnaire that used to evaluate this system for 35 participants. Based on the questionnaire results, more than half number of participants or approximately 51.4% have never used any tourism application. When the participants used this application, over 94.3% agree that scan landmark, view events, emergency, and history features in the application are really useful for their travelling. However, due to some problems that occur during testing, such as unstable internet connection, confusing interface, application crash, etc., only 87.5% of participants admitted that the application is easy to use. For the main feature, i.e., scan landmark, 94.3% participants have experienced that the feature already correctly recognized any tested tourist attractions and
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
562
7
Table 2: The results of white box testing on the backend of smart travelling application
Menu Login User City Landmark Event Logout
Function Description Login Search users based on the user id, role, name, email, or phone Update and delete user data Search city data based on city name Add, update, and delete city data Search based on landmark name Add, update, and delete landmark data Search any events based on city, event date, and event name Add, update, and delete event data Logout
Result OK OK OK OK OK OK OK OK OK OK
Evaluation Questionnaire 1.
2.
3.
4.
5.
6.
7.
8.
9.
Have you used any tourism application?
○
Yes
○
No
Does the smart travelling help you in recognizing tourist attractions?
○
Yes
○
No
Did the smart travelling already correctly recognize tourist attractions?
○
Yes
○
No
Does the provided event information feature help you?
○
Yes
○
No
Does the emergency menu help you in finding the nearest hospital and/or police office?
○
Yes
○
No
Does the history menu facilitate you for reminding the recognized attractions?
○
Yes
○
No
Is the smart travelling easy to use?
○
Yes
○
No
What is the problem while you test the application?
Do all features work properly according to their functions?
○
Yes
○
No
10. In your opinions, which feature is the most helpful for your travel?
☺ Thank You ☺
Fig. 8: Questionnaire form for evaluating the Smart Travelling application
the rest still have some problem with it. The most powerful feature in this application based on the participants choice is the scan landmark which can automatically recognize the tourist attractions. In addition to this application evaluation, we also perform an experiment for the scan landmark feature. This experiment focuses only on the image recognition accuracy. We used 50 images as the testing data which are collected from Google Images. Those images are not constrained in uniform resolution and only contain Indonesia’s tourism
8
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
563
attractions. By performing this experiment, we achieved 86% accuracy which calculated using Eq. 1. Fig. 9 presents the correct identification examples that have been obtained during this experiment.
Accuracy =
# correctly recognized ∗ 100% # total testing data
(1)
Fig. 9: Example of the correct recognition in scan landmark feature
5. Conclusion and Future Works In this paper, we present the study of constructing an android application for tourism in Indonesia. The application is named Smart Travelling and has many features in order to facilitate the tourists who visit Indonesia. Based on the conducted study, the smart travelling application is successfully published in Google Play Store and evaluated by 35 selected participants. It shows the smart travelling application is really useful for tourists either local or foreigner, especially the image recognition feature which can automatically recognize the tourist attractions with great performance, i.e., 86% accuracy. Even though the application already launched, there are several points which can be considered for the improvement in the future, such as use our own algorithm for the image recognition instead of Google Cloud Vision API, reconstruct the application for other platforms, redesign the interface, etc. References
1. Pariwisata, K.. Perkembangan WISNAS (Wisatawan Nasional) Tahun 2011-2016. http://www.kemenpar.go.id/userfiles/rekap%20wisnas.pdf; 2016. Accessed: 2017-01-20. 2. Pariwisata, K.. Perkembangan Wisatawan Mancanegara Tahun 2011-2015. http://www.kemenpar.go.id/userfiles/rekap%20wisman.pdf; 2015. Accessed: 2017-01-20. 3. Platform, G.C.. Cloud Vision API Features. https://cloud.google.com/vision/; 2017. Accessed: 2017-02-27. 4. McSpadden, K.. Infographic: Android dominates Southeast Asia’s smartphones. https://e27.co/infographic-android-dominates-southeast-a 2015. Accessed: 2017-01-22. 5. Tobing, R.D.H.. Mobile tourism application for samosir regency on android platform. In: Technology Management and Emerging Technologies (ISTMET), 2015 International Symposium on. IEEE; 2015, p. 249–253. 6. Chao, C.L., Lin, K.C., Tseng, C.M., Erdene, T., Chen, Y.F.. An app for promoting health and local tourism. In: Computer, Consumer and Control (IS3C), 2014 International Symposium on. IEEE; 2014, p. 1175–1178. 7. Tian, F., Xu, F., Fu, J.. Augmented reality technology overview for tourism app development. In: Machine Learning and Cybernetics (ICMLC), 2013 International Conference on; vol. 4. IEEE; 2013, p. 1483–1489. 8. Bandung, Y., Hendra, Y.F., Subekti, L.B.. Design and implementation of digital signage system based on raspberry pi 2 for e-tourism in indonesia. In: Information Technology Systems and Innovation (ICITSI), 2015 International Conference on. IEEE; 2015, p. 1–6. 9. Ostrand, T.. Encyclopedia of Software Engineering. John Wiley & Sons, Inc.; 2002.
Available online at www.sciencedirect.com
ScienceDirect
Procedia Computer Science 00 (2017) 000–000 Procedia Computer Science (2017) 000–000 Procedia Computer Science 11600 (2017) 556–563
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
2nd International Conference on Computer Science and Computational Intelligence 2017, 2nd International Conference on Computer Science andBali, Computational ICCSCI 2017, 13-14 October 2017, Indonesia Intelligence 2017, ICCSCI 2017, 13-14 October 2017, Bali, Indonesia
Mobile Mobile Smart Smart Travelling Travelling Application Application For For Indonesia Indonesia Tourism Tourism Meiliana, Devita Irmanti, Mochamad Rizal Hidayat, Nur Viani Amalina, Dewi Suryani∗∗ Meiliana, Devita Irmanti, Mochamad Rizal Hidayat, Nur Viani Amalina, Dewi Suryani
Computer Science Department, School of Computer Science, Bina Nusantara University, Jl. K. H. Syahdan No. 9, DKI Jakarta, 11480, Indonesia Computer Science Department, School of Computer Science, Bina Nusantara University, Jl. K. H. Syahdan No. 9, DKI Jakarta, 11480, Indonesia
Abstract Abstract Indonesia is a country that has seventeen thousand islands and located in a very strategic location. That reason makes Indonesia Indonesia a country that has seventeen thousand and located a very location. That makes of Indonesia becoming is a place that is interesting to be visited byislands many tourists eitherinlocal or strategic international. Every yearreason the number tourists becoming a place that is interesting to be visited by many tourists either local or international. Every year the number of tourists 1,2 The who visited Indonesia has greatly increased 1,2 and transformed the tourism to be an important sector for Indonesia economical. and transformed the tourism to be an important sector for Indonesia economical. The who visited Indonesia has greatly increased rapid development of tourism sector in Indonesia must be equilibrated by the evolution of technology and the supported facilities. rapid tourism sector in Indonesia be important equilibrated by the technology the activities. supportedBy facilities. Now, development the usage of aofsmartphone technology is onemust of the things andevolution a part ofof modern peopleand daily taking Now, the usage of of aa smartphone, smartphone technology one of thetoimportant things and aapplication part of modern activities. By taking the advantages this study isisconducted develop an android that ispeople nameddaily Smart Travelling. Smart the advantages of a smartphone, this study is conducted to develop an android application that is named Smart Travelling. Smart travelling has many features that can facilitate every tourist who visited Indonesia, such as recognizes any tourist attractions, shows travelling features can facilitate everythe tourist who visited Indonesia, such asfor recognizes any tourist shows any eventshas thatmany nearby visitedthat attractions, displays nearest police station and hospital an emergency case, attractions, and saves the hisany that nearby objects. visited attractions, displays nearest police station and hospital an emergency case, and saves theusing history events of the recognized The main idea of thethe tourist attractions recognition here is for to implement an image recognition tory of the recognized objects. The main idea of the tourist attractions recognition here is to implement an image recognition using 3 landmark detection feature from Google Cloud Vision Application Program Interface (API) 3 technology which helps the tourist landmark detection and feature from Cloudregarding Vision Application Interface technology helpsapplication the tourist to easily remember access anyGoogle information the visited Program attractions. In this (API) study, the result is awhich completed to easily remember and access any information regarding the visited attractions. In this study, the result is a completed application which is evaluated using questionnaires to 35 randomly selected participants. Based on the evaluation, the smart travelling achieves which evaluated questionnaires to 35agree randomly selected evaluation, smart travelling achieves positiveis inputs andusing mostly the participants that the imageparticipants. recognition Based featureonis the really helping the them. Furthermore, we also positive inputs and mostly the participants agree that the image recognition feature is really helping them. Furthermore, we also obtained a significant result for the scan landmark testing, which is 86% for the accuracy. obtained a significant result for the scan landmark testing, which is 86% for the accuracy. c 2017 2017 The The Authors. Authors. Published Published by by Elsevier Elsevier B.V. © B.V. c 2017 The Authors. Published by Elsevier B.V. Peer-review Computer Science Science and and ComPeer-review under under responsibility responsibility of of the scientific committee of the 2nd International Conference on Computer Peer-review under responsibility putational Intelligence 2017.2017.of the scientific committee of the 2nd International Conference on Computer Science and ComComputational Intelligence putational Intelligence 2017. Keywords: Image Recognition; Android application; Google Cloud Vision; Tourism; Keywords: Image Recognition; Android application; Google Cloud Vision; Tourism;
1. Introduction 1. Introduction Nowadays, the outgrowth of technologies and globalization era demand ease and automation in various aspects Nowadays, outgrowth of technologies and demand economic ease and automation in on various aspects of human life, the including tourism. In Indonesia, theglobalization tourism is anera important sector. Based the ministry of human life, including 1 tourism. In Indonesia, the tourism is an important economic sector. Based on the ministry of tourism data in 2016 1 , the number of local tourists in 2016 has been growing exponentially with over 6.6 million of tourism datahad in 2016 , the 1.97% numbercomparing of local tourists 2016 has beenFurthermore, growing exponentially over that 6.6 million people, which increased to the inprevious years. the foreignwith tourists visited people, which had increased 1.97% comparing to the previous years. Furthermore, 2 the foreign tourists that visited Indonesia had also increased 10.29% or approximately 10.4 million people in 2015 2 . Indonesia had also increased 10.29% or approximately 10.4 million people in 2015 . ∗ ∗
Corresponding author. Tel.: +62-21-534-5830 ext 2188; fax: +62-21-530-0244 Corresponding Tel.: +62-21-534-5830 ext 2188; fax: +62-21-530-0244 E-mail address:author. [email protected] E-mail address: [email protected]
c 2017 The Authors. Published by Elsevier B.V. 1877-0509 cunder 1877-0509 2017 The Authors. of Published by Elsevier B.V. the 2nd International Conference on Computer Science and Computational IntelliPeer-review the scientific committee 1877-0509 © 2017responsibility The Authors. Published Elsevier of B.V. Peer-review responsibility of the scientificbycommittee of the 2nd International Conference on Computer Science and Computational Intelligence 2017. under Peer-review under responsibility of the scientific committee of the 2nd International Conference on Computer Science and gence 2017. Computational Intelligence 2017. 10.1016/j.procs.2017.10.059
2
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
557
The rapid development of tourism sector in Indonesia must be equilibrated by the evolution of technology and the supported facilities. Now, the usage of a smartphone technology is one of the important things and a part of modern people daily activities. The utilities of the smartphone are not only in telecommunication but also in various fields, such as looking for information, entertainment, work supporter, data storage, photography, guidance, etc. By taking the advantages of a smartphone, this study is conducted to develop a mobile application. Currently, there are many kinds of platforms installed in a smartphone and the most popular platforms used are Android, iOS, and Windows Phone. Based on the number of smartphone users in 6 ASEAN countries 4 , most of the users used smartphones Android-based. Especially in Indonesia, at 2015, the Android users had reached 41 million users compared to 2.8 million iOS users. Therefore, this application is built in the Android smartphone. Indonesia is a country that is located mainly in South-east Asia and consists of seventeen thousand islands. Its strategic location makes Indonesia has many interesting sights to be visited by tourists both local and foreign. However, there is still not many applications that specific for all tourist attractions in Indonesia. In this paper, we report a tourism mobile application for Indonesia, which is named ”Smart Travelling” and is built only for the Android platform. The smart travelling has several functionalities, such as recognizes any tourist attractions, shows any events that nearby visited attractions, displays the nearest police station and hospital for the emergency case, and saves the history of the recognized objects. Furthermore, this application has a special feature that is not available in other similar applications. The feature is an image recognition that can be used to automatically recognize an object and give any information about it. The main purpose of this application is to help tourists to recognize any tourist attractions in Indonesia and get all related information. The remaining of this paper is organized as follows. In Section 2, we describe the relation of other studies for tourism purpose to ours. Next, all features in the proposed application and its methodology will be discussed in Section 3. Section 4 explains how we evaluate the application and its results. Finally, Section 5 concludes this paper and discusses the future development. 2. Related Works Tourism is an essential economic sector in a country. By focusing on tourism, many advantages can be achieved, such as an improvement in various sectors especially provision of employment, income, production, etc. Due to the necessity of tourism, several applications are built in order to endorse its process 5,6,7,8 . The work of Tobing 5 is one of the studies that conducts an android application for tourism in Samosir Regency, North Sumatera-Indonesia. It mainly focuses on the implementation of Google Map Application Program Interface (API) and integrated social media. Similar to Tobing 5 , Chao 6 developed the android application by integrating global positioning system (GPS). However, Chao 6 concentrated on Taiwan tourism and folk-cultural education instead of Indonesia tourism. Furthermore, Tian et al. 7 proposed an augmented reality technology in developing tourism application that can be used on any mobile platforms. In contrast to the other approaches, Bandung 8 implements digital signage system and Raspberry Pi 2 Model B as its client for supporting e-tourism in Indonesia. This study proposed an application for presenting Indonesia tourism with special features. Like Tobing 5 and Chao 6 , this application runs on the android platform. However, we focused on tourism area for entire Indonesia, which is a generalization of Chao 6 ’s approach. Unlike Tian et al. 7 , we applied the image recognition technique instead of augmented reality technology. Compared to the existing similar applications such as Pesona Indonesia and Wisata Lokal Travel Indonesia, our application has the combinations of all their features and offered a special feature which has not implemented in the others. The special feature here means the image recognition that can automatically recognize any tourist attractions which had taken by the tourist. Furthermore, our application also has a comprehensive tutorial which definitely helps the tourist to access and use our application in the proper way, and it is not limited in several Indonesia areas only. 3. System Overview As mentioned in the previous sections, the purpose of this study is to construct a mobile application for facilitating any tourists who are visiting Indonesia. The application is namely Smart Travelling and developed in Android platform. It has several main features that described in Fig. 1 and will be detailed as follows.
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
558
3
Fig. 1: Main menu of Smart Travelling contains scan landmark, event, emergency, history, tutorial, and profile features
3.1. Scan Landmark This feature has implemented the image recognition technology using Google Cloud Vision Application Program Interface (API) 3 . Google Cloud Vision API allows the developer to extract data from an image by encapsulating the machine learning model using REST API. It can classify any images into categories, detect individual objects, and read any printed words of an image. The Google Cloud Vision API can analyze either the uploaded images or the integrated images of the gallery in the Google Cloud Storage. There are many functionalities that are provided by the Google Cloud Vision API, however, this menu only applied the landmark detection feature, which identifies a natural or artificial object in an image. This feature not only focuses on recognizing the tourist attractions but also provides the additional information about it by automatically using Google search when the user requested as described in Fig. 2.
Fig. 2: (First) Scan landmark menu when the user selected it. (Second) Sub menu of scan landmark for performing image recognition using the camera. (Third) Sub menu of scan landmark for performing image recognition in the gallery. (Fourth) Scan landmark menu redirects to Google search engine when the user has chosen to search information regarding the recognized object
The scan landmark feature is the main focus of this study which takes the advantages of the landmark detection of Google Cloud Vision API. In this landmark detection, the first thing is getting the input images either via camera or by the gallery. Afterwards, the application sends a request to the Google Vision API for detecting the object. When the request is successful, the Google Vision API server returns the identifying information such as an entity ID, the landmark’s name, the landmark’s location, and the rectangle for showing the recognized object in the input image. However, the information that shown for the user is only the landmark’s name as shown in Fig. 9.
4
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
559
3.2. Event This feature provides any information about all ongoing or upcoming events in Indonesia which are gotten from the database that implemented using MySQL. Events data in the database were inputted by admin using a web-based application as the user interface and SQL code to manipulate the data. After the data is inputted, all events are viewed in Event Menu at the mobile version application as shown in Fig. 3. The user can view the detail location of the desired event based on Google Map API coordinate which is stored in the database too. There are also searching option to look for the event based on name or city which the application retrieve it from the user input and search the database using SQL command.
Fig. 3: Event menu while searching an event
3.3. Emergency This feature provides the nearest public services that users need immediately. The program reads the Android GPS current location and then search the nearest location of hospital or police station in radius 5 km using Google Maps API. It is depicted in Fig. 4. After the nearest hospital or police station is found, Google Maps API will return all coordinates of desired public service place and the fastest route to the position which will be displayed by the application.
Fig. 4: (Left) View of the emergency menu when the user selected it. (Center) The map displays all hospital’s coordinates after the user selected hospital sub menu. (Right) Map shows all police station’s coordinates when police station menu is selected
560
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
5
3.4. History The application implements a history feature which is used to view any landmark that has been scanned using the scan landmark menu. Everytime the user scans landmark, the landmark information will be stored into MySQL database. The data then will be displayed by the application using SQL code into History menu which already sorted by the latest scanned date. Fig. 5 displays the view of history menu when this menu is chosen.
Fig. 5: View of history menu when the user selected the menu
In addition, this application also provides tutorial and profile features. The tutorial feature displays all step procedures of each menu in the proposed application user interface in order to simplify the utilization of the application. The tutorial will provide all information for the user to use the application and described in Fig. 6.
Fig. 6: Tutorial menu for scan landmark, event, emergency, and history
Moreover, users can manage their own profile in the application through the profile menu, which is depicted in Fig. 7. The detail information of the user can easily be updated, such as name, email address, and phone number. Their password, which is used to login to the application, can also be changed. The application will update the new data of users to MySQL database using SQL command once the user confirms the changes. 4. Evaluation In order to evaluate the usage of the application and the user satisfactory of the smart travelling, this study has conducted white box and questionnaire evaluations after completing the application development process. White box testing is an evaluation method that depends on the internal structure of the application and executed during actual system operation 9 . Here, the white box evaluation will be done not only for front-end system but also for the back-end
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
6
561
Fig. 7: (Left) Profile menu in Smart Travelling. (Center) Update profile sub menu. (Right) Change password sub menu
component that had implemented in the web-based application. The results of these white box evaluations are depicted in Table 1 and Table 2. Table 1: The results of white box testing on the front-end of smart travelling application
Menu Register Login Scan Landmark Emergency History Tutorial Event Profile Logout
Function Description Register Login Scan landmark from camera and gallery Get tourist attraction name Search additional information of the attraction in Google Add the recognized object data into history menu Perform the emergency navigation Show the nearest location for emergency View the history data of recognized tourist objects Search additional information about the object in Google View information how to use the application View all ongoing and upcoming events information in Indonesia Search event based on the event name or the visited city View and update user data Change password Logout
Result OK OK OK OK OK OK OK OK OK OK OK OK OK OK OK OK
For the questionnaire evaluation, we provide 10 questions with the total number of participants is 35 people, which is randomly selected. Before filling the questionnaire, the participants need to install the application on their smartphones. At first, the user needs to download the smart travelling application in the Play Store. It can be done in two ways, i.e., search by keyword ”Smart Travelling” or click this URL ”https://play.google.com/store/apps/details?id=com.skripsi.smarttravelling”. Afterwards, install it and wait until the process is finished, then the application can be tested. Fig. 8 shows the questionnaire that used to evaluate this system for 35 participants. Based on the questionnaire results, more than half number of participants or approximately 51.4% have never used any tourism application. When the participants used this application, over 94.3% agree that scan landmark, view events, emergency, and history features in the application are really useful for their travelling. However, due to some problems that occur during testing, such as unstable internet connection, confusing interface, application crash, etc., only 87.5% of participants admitted that the application is easy to use. For the main feature, i.e., scan landmark, 94.3% participants have experienced that the feature already correctly recognized any tested tourist attractions and
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
562
7
Table 2: The results of white box testing on the backend of smart travelling application
Menu Login User City Landmark Event Logout
Function Description Login Search users based on the user id, role, name, email, or phone Update and delete user data Search city data based on city name Add, update, and delete city data Search based on landmark name Add, update, and delete landmark data Search any events based on city, event date, and event name Add, update, and delete event data Logout
Result OK OK OK OK OK OK OK OK OK OK
Evaluation Questionnaire 1.
2.
3.
4.
5.
6.
7.
8.
9.
Have you used any tourism application?
○
Yes
○
No
Does the smart travelling help you in recognizing tourist attractions?
○
Yes
○
No
Did the smart travelling already correctly recognize tourist attractions?
○
Yes
○
No
Does the provided event information feature help you?
○
Yes
○
No
Does the emergency menu help you in finding the nearest hospital and/or police office?
○
Yes
○
No
Does the history menu facilitate you for reminding the recognized attractions?
○
Yes
○
No
Is the smart travelling easy to use?
○
Yes
○
No
What is the problem while you test the application?
Do all features work properly according to their functions?
○
Yes
○
No
10. In your opinions, which feature is the most helpful for your travel?
☺ Thank You ☺
Fig. 8: Questionnaire form for evaluating the Smart Travelling application
the rest still have some problem with it. The most powerful feature in this application based on the participants choice is the scan landmark which can automatically recognize the tourist attractions. In addition to this application evaluation, we also perform an experiment for the scan landmark feature. This experiment focuses only on the image recognition accuracy. We used 50 images as the testing data which are collected from Google Images. Those images are not constrained in uniform resolution and only contain Indonesia’s tourism
8
Meiliana et al. / Procedia Computer Science 116 (2017) 556–563 Meiliana et al. / Procedia Computer Science 00 (2017) 000–000
563
attractions. By performing this experiment, we achieved 86% accuracy which calculated using Eq. 1. Fig. 9 presents the correct identification examples that have been obtained during this experiment.
Accuracy =
# correctly recognized ∗ 100% # total testing data
(1)
Fig. 9: Example of the correct recognition in scan landmark feature
5. Conclusion and Future Works In this paper, we present the study of constructing an android application for tourism in Indonesia. The application is named Smart Travelling and has many features in order to facilitate the tourists who visit Indonesia. Based on the conducted study, the smart travelling application is successfully published in Google Play Store and evaluated by 35 selected participants. It shows the smart travelling application is really useful for tourists either local or foreigner, especially the image recognition feature which can automatically recognize the tourist attractions with great performance, i.e., 86% accuracy. Even though the application already launched, there are several points which can be considered for the improvement in the future, such as use our own algorithm for the image recognition instead of Google Cloud Vision API, reconstruct the application for other platforms, redesign the interface, etc. References
1. Pariwisata, K.. Perkembangan WISNAS (Wisatawan Nasional) Tahun 2011-2016. http://www.kemenpar.go.id/userfiles/rekap%20wisnas.pdf; 2016. Accessed: 2017-01-20. 2. Pariwisata, K.. Perkembangan Wisatawan Mancanegara Tahun 2011-2015. http://www.kemenpar.go.id/userfiles/rekap%20wisman.pdf; 2015. Accessed: 2017-01-20. 3. Platform, G.C.. Cloud Vision API Features. https://cloud.google.com/vision/; 2017. Accessed: 2017-02-27. 4. McSpadden, K.. Infographic: Android dominates Southeast Asia’s smartphones. https://e27.co/infographic-android-dominates-southeast-a 2015. Accessed: 2017-01-22. 5. Tobing, R.D.H.. Mobile tourism application for samosir regency on android platform. In: Technology Management and Emerging Technologies (ISTMET), 2015 International Symposium on. IEEE; 2015, p. 249–253. 6. Chao, C.L., Lin, K.C., Tseng, C.M., Erdene, T., Chen, Y.F.. An app for promoting health and local tourism. In: Computer, Consumer and Control (IS3C), 2014 International Symposium on. IEEE; 2014, p. 1175–1178. 7. Tian, F., Xu, F., Fu, J.. Augmented reality technology overview for tourism app development. In: Machine Learning and Cybernetics (ICMLC), 2013 International Conference on; vol. 4. IEEE; 2013, p. 1483–1489. 8. Bandung, Y., Hendra, Y.F., Subekti, L.B.. Design and implementation of digital signage system based on raspberry pi 2 for e-tourism in indonesia. In: Information Technology Systems and Innovation (ICITSI), 2015 International Conference on. IEEE; 2015, p. 1–6. 9. Ostrand, T.. Encyclopedia of Software Engineering. John Wiley & Sons, Inc.; 2002.