Web-based Safari Review System Development using Microblog Analyzed Data

• Material Type: 博士論文
• Title: Web-based Safari Review System Development using Microblog Analyzed Data
• Author/Editor: Victor Alex Silaa
• Author Heading: Victor Alex Silaa
• Publication Date: 2023-09
• Publication Date (W3CDTF): 2023-09
• Alternative Title: マイクロブログの解析データを利用したWebベースのサファリレビューシス テム開発
• Degree grantor/type: 北見工業大学
• Date Granted: 2023-09-05
• Date Granted (W3CDTF): 2023-09-05
• Dissertation Number: 甲第211号
• Degree Type: 博士（工学）
• Text Language Code: eng
• Target Audience: 一般
• Note (General): In this study, I propose the use of online microblogs as review supplements and demonstrate theirapplicability through a designed tourist support system that aims to provide additional opinionsand up-to-date points of interest to the less-known tourist spots. In realizing this proposal, I useInformation Extraction (IE), Artificial Intelligence (AI), and Natural Language Processing (NLP)- based techniques. The proposed approach folds into three.First, through the use of geotagged tweets. Tweets that contain geolocation informationare considered geotagged and therefore treated as possible tourist on-spot opinions. The mainchallenge, however, is to confirm the authenticity of the extracted tweets. This stage includes theuse of location clustering and classification techniques. Specifically, extracted geotagged tweetsare clustered by using location information and then annotated taking into consideration specificfeatures applied to machine learning-based classification techniques. As for the machine learning(ML) algorithms, I adopt a fine-tuned transformer neural network-based BERT model whichimplements the information of token context orientation for better classification.Second, I studied geolocatability of ungeotagged tweets so that they can be used as reviewalternatives. Ungeotagged tweets have no geolocation information attached so it is difficult toassociate with specific location. Furthermore, Twitter data is typically noisy and consists ofungrammatical or informal phraseology and non-standard vocabulary, which additionally causesthe feature sparsity problem, resulting in low classifier performance.To address this, I proposed the use of a two-stage process, a transformer-based model for theclassification of primary tweets, and a combination of impact words like location mention or eventmention for location inferring. Additionally, I evaluate a range of pre-processing techniques for textcategorization to accurately obtain a proper set that collectively contributes to the improvementof prediction accuracy. A classification framework created here relies on a fine-tuned transformerneural network model which learns from tweet contents and predicts the locations from which thosetweets were sent - with a limited application in the detection of widely known general locations- such as tourist spots. I learned that the average 0.84 F1 score of a pre-trained DistilBERTlanguage model outperformed other tested models when tested on different pre-processing datasets.Furthermore, i evaluated the effect of impact words like location mention, and event mention onthe geolocation estimation, and model accuracy improvement when impact words are involved orremoved. To investigate the effect of impact words on a classification model, i first computedthe weighting of words using TFIDF and futher created a likelihood wordlist. I discovered modelaccuracy improvement as much as 6% when impact words are involved compared to when theyare removed which suggests positive influence of impact words in geolocatability. I also discoveredwrong weighted impact words that negatively contributes to the model performance and byeliminating them, the model F1 score improved by 3%.Third, I demonstrate the applicability of these two approaches by designing a tourist supportsystem and mapping extracted opinions to their respective tourist spots as touristic information.
• DOI: 10.19000/0002000492; https://doi.org/10.19000/0002000492
• Persistent ID (NDL): info:ndljp/pid/13116907; https://dl.ndl.go.jp/pid/13116907
• Collection: 障害者向け資料
• Collection (Materials For Handicapped People:1): テキストデータ
• Collection (particular): 国立国会図書館デジタルコレクション > デジタル化資料 > 博士論文; https://dl.ndl.go.jp/collections/A00014
• Acquisition Basis: 博士論文（自動収集）
• Date Accepted (W3CDTF): 2023-12-05T21:40:42+09:00
• Format (IMT): application/pdf
• Access Restrictions: 国立国会図書館内限定公開
• Service for the Digitized Contents Transmission Service: 図書館・個人送信対象外
• Availability of remote photoduplication service: 可
• Periodical Title (URI): http://dx.doi.org/10.19000/0002000492; https://kitami-it.repo.nii.ac.jp/records/2000492
• Data Provider (Database): 国立国会図書館 : 国立国会図書館デジタルコレクション; https://dl.ndl.go.jp

• Summary, etc.: In this study, I propose the use of online microblogs as review supplements and demonstrate their applicability through a designed tourist support sys...