scholarly journals Construction of Machine Learning Model Based on Text Mining and Ranking of Meituan Merchants

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Yin Tang ◽  
Dongxue Liao ◽  
Shuqiang Huang ◽  
Qing Fan ◽  
Liang Liu

In the Web 2.0 era, the problem of uneven quality and overload of online reviews is very serious, and the cognitive cost of obtaining valuable content from them is getting higher and higher. This paper explores an effective solution to address comment overload by means of information recommendation in order to improve the utilization of online information and information service quality. This paper proposes a review ranking recommendation scheme that focuses on the information quality of reviews and places more emphasis on satisfying users’ personal information need. The paper’s approach is used to extract and rank low-frequency keywords that appear only once in the comment set. The more useful the extracted phrases are, the more useful this review will be and the higher the usefulness votes will be, which can reflect the actual situation of this product more objectively and accurately and facilitate better consumption decisions for consumers. The experimental results show that users’ satisfaction with the perceived usefulness of the reviews is jointly influenced by the information quality of Meituan’s reviews and users’ individual information needs; the recommendation strategy achieves the organic integration of the two, and the evaluation results under three different recommendation modes show that compared with “interest recommendation” and “utility recommendation,” the satisfaction score of “fusion recommendation” is the highest

2021 ◽  
Author(s):  
Anagha Kulkarni ◽  
Mike Wong ◽  
Tejasvi Belsare ◽  
Risha Shah ◽  
Diana Yu Yu ◽  
...  

BACKGROUND The Internet has become a major source of health information especially for adolescents and young adults. Unfortunately, inaccurate, incomplete or outdated health information is widespread online. Often adolescents and young adults turn to authoritative websites such as the student health center (SHC) website of the university they are attending to obtain reliable health information. Although most on-campus SHC clinics comply with the American College Health Association (ACHA) standards, their websites are not subject to any standards or code of conduct. In the absence of quality standards or guidelines, the monitoring and compliance processes do not exist for SHC websites either. As such, there is no oversight on the health information published on the SHC websites by any central governing body. OBJECTIVE Our objective is to enable researchers to monitor online information quality at scale. We have created a tool that can efficiently quantify the quality of information posted on SHC websites about a health topic. Specifically, this quantitative tool provides information on quality, such as reading ease, coverage of the topic, and the degree of fact-based objective information. METHODS Our cross-functional team has designed and developed an open-source software, QMOHI: Quantitative Measures of Online Health Information, using the Agile software development methodology. The QMOHI tool finds the SHC website and gathers information on the specific health topic of interest from a prespecified list of university websites. Based on the retrieved text, the tool computes eight different quality metrics. The QMOHI tool is a fully automated tool that is designed to be scalable, generalizable, and robust. RESULTS The first empirical evaluation shows that the QMOHI tool is highly scalable and substantially more efficient than the manual approach of assessing online information quality. The second experimental results demonstrate QMOHI’s ability to work effectively with starkly different health topics (COVID, Cancer, LARC, and Condom) and with narrowly focused topics (hormonal IUD and copper IUD); thereby establishing the generalizability and versatility of the tool. The results from the last experiment demonstrate that QMOHI is not vulnerable to typical structural changes that SHC websites may undergo (e.g. URL changes) over a long period of time. QMOHI is able to support longitudinal studies by being robust to such website changes. CONCLUSIONS QMOHI allows public health researchers and practitioners to conduct large-scale studies of SHC websites that were previously too time intensive. The capability to generalize broadly or focus narrowly allows for wide applications of QMOHI, equipping researchers to study both mainstream and underexplored health topics. QMOHI’s ability to robustly analyze SHC websites periodically facilitates longitudinal investigations and monitor SHC progress. QMOHI serves as a launching pad for our future work that aims to develop a broadly applicable public health tool for online health information studies with potential applications far beyond SHC websites.


Prologia ◽  
2021 ◽  
Vol 5 (1) ◽  
pp. 127
Author(s):  
Julian Andrew ◽  
Rezi Erdiansyah

As the people's shopping habits via online starts to emerge, the e-commerce industry in Indonesia has also developed. In 2018, it was noted that 11.9% of Indonesian people were shopping online. However, in the midst of the vastness of online platforms with millions of items found in online storefronts, consumers need more information as their reference to arousing buying interest. As one of the biggest e-commerce players in Indonesia, Tokopedia provides features that enable sellers and consumers to exchange information regarding the items. In Tokopedia, prospective buyers can see electronic word of mouth messages, online reviews, and other additional information about the items that are known to be very influential in generating buying interest. This study uses a quantitative approach with an explanatory type in which the research seeks to find the effect of electronic word of mouth, online review, and the quality of information on buying interest of Jakarta students in Tokopedia e-commerce. The data collection technique used was purposive sampling by distributing questionnaires to 100 samples via online. Based on this research’s results, it was found that electronic word of mouth, online review, and information quality affect buying interest of students in Jakarta by 46% while the other 54% were influenced by other factors not examined in this study.Seiring dengan munculnya kebiasaan berbelanja masyarakat melalui online, industri e-commerce di Indonesia pun turut berkembang. Pada tahun 2018, tercatat bahwa sebanyak 11,9% orang di Indonesia melakukan kegiatan belanja secara online. Namun, di tengah luasnya platform online dengan jutaan barang yang terdapat di etalase online membuat konsumen membutuhkan informasi yang lebih sebagai bahan referensi dalam menimbulkan minat beli. Tokopedia merupakan pelaku e-commerce terbesar di Indonesia menyediakan fitur-fitur yang memungkinkan penjual dan konsumen untuk menulis dan bertukar informasi seputar barang tersebut. Di dalam Tokopedia, para calon pembeli dapat melihat pesan electronic word of mouth, online review, dan informasi-informasi tambahan lainnya seputar barang-barang yang dijual yang diketahui sangat berpengaruh dalam memunculkan minat beli. Penelitian ini menggunakan pendekatan kuantitatif dengan jenis eksplanatif dimana penelitian berusaha menemukan pengaruh e-WOM, kualitas informasi, dan online review terhadap minat beli mahasiswa Jakarta pada e-commerce Tokopedia. Teknik pengumpulan data yang digunakan adalah purposive sampling dengan menyebarkan kuesioner kepada 100 sampel secara online. Penelitian menemukan bahwa electronic word of mouth, online review, dan kualitas informasi berpengaruh terhadap minat beli pada mahasiswa di Jakarta sebanyak 46%, sedangkan 54% dipengaruhi oleh faktor-faktor lain yang tidak diteliti pada penelitian ini.


Author(s):  
Tiana Tiana ◽  
Bambang Sugiharto ◽  
Indah Umiyati

This study aims to test and analyze the quality of the system against perceived usefulness, the quality of information against perceived usefulness, the quality of the system to user satisfaction, Information Quality to user satisfaction and perceived usefulness to user satisfaction. The population in this research is employees teller and customer service department of Bank BJB in subang city. Sampling technique is nonprobability sampling by using purposive sampling method. The type of data used is primary data with questionnaire data collection method. Hypothesis testing used is Structural Equation Modeling (SEM) approach with variance Structural equation test or better known as Partial Least Square (PLS). The result of the research shows that Quality System has a significant positive effect on Perceived Usefulness, Quality of Information has a significant positive effect on Perceived Usefulness, Quality of System has a significant positive effect on user satisfaction, Information Quality has positive significant effect on user satisfaction, and Perceived Usefulness has no effect to User Satisfaction


2019 ◽  
Vol 3 (2) ◽  
pp. 288-299
Author(s):  
Winanda Wahana Wargadalam

Nowadays technology grow extremely fast and it is helpful in communication and transactions, this effect to human behavior that people consider using technology intentively in daily activites cause of its facilities, one of facilities is digital economic transactions or e-money. Due to increasing of technology, there are some companies run their business form e-money based applications, one of the applications is the PayTren. The purposed of this research is to examine the satisfactory of the user who apply PayTren applications, such as the influence of information quality, system quality, service quality and security. The method of the research is based on DeLone and McLean models, then it will be modified. The research’s data is quantitative data. researchers collected 89 questionnaires from the population of PayTren application users in Batam by using a purposive sampling technique.  The conclusion of the research show that there is an influence between the quality of information, service quality and security on the satisfaction of e-money users based on the PayTren application, while it is found that there is no influence between the quality of the system on the satisfaction of e-money users based on PayTren applications.


2003 ◽  
Vol 42 (02) ◽  
pp. 134-142 ◽  
Author(s):  
P. Doupi ◽  
Jeroen van den Hoven ◽  
K. Lampe

Summary Objectives: Quality of online health resources remains a much debated topic, despite considerable international efforts. The lack of a systematic and comprehensive conceptual analysis is hindering further progress. Therefore we aim at clarifying the origins, nature and interrelations of pertinent concepts. Further, we claim that quality is neither a necessary nor a sufficient condition for Internet health resources to produce an effect offline. As users’ trust is also required, we examine the relation of quality aspects to trust building online. Methods: We reviewed and analyzed the key documentation and deliverables of quality initiatives, as well as relevant scientific publications. Using the insights of philosophy, we identified the elementary dimensions which underlie the key concepts and theories presented so far in the context of online health information quality. We examined the interrelations of various perspectives and explored how trust as a phenomenon relates to these dimensions of quality. Results: Various aspects associated with the quality of online health resources originate from four conceptual dimensions: epistemic, ethical, economic and technological. We propose a conceptual framework that incorporates all these perspectives. We argue that total quality exists only if all four dimensions have been addressed adequately and that high total quality is conducive to warranted trust. Conclusions: Quality and trust are intertwined, but distinct concepts, and their relation is not always straightforward. Ideally, trust should track quality. Apprehending the composition of these concepts will help to understand and guide the behavior of both users and providers of online information, as well as to foster warranted trust in online resources. The framework we propose provides a conceptual starting point for further deliberations and empirical work.


2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Laila Al-Omairi ◽  
Hosam Al-Samarraie ◽  
Ahmed Ibrahim Alzahrani ◽  
Nasser Alalwan

PurposeIn any developing country, the critical benefits for ensuring proper utilization of e-government learning services have basically emerged from the integration of underlying processes – not only across different levels of government but also across different functions and agencies. The literature showed a notable lack of research to underline the limited use of e-government learning services in a university setting. Therefore, this study was conducted to examine students' intention to adopt e-government learning services in a developing country. Precisely, the existent to which reputation, social influence, information quality and system quality of e-government learning services would influence students' trust, perceived usefulness and perceived learning values was investigated.Design/methodology/approachThe effect from these variables on students' intention to use e-government learning services was also investigated. A total of 406 questionnaires were collected and analyzed using partial least squares (PLS) modeling.FindingsThe results showed varying relationships between the study's variables. It is anticipated that understanding these relationships can predict future learning trends of e-government learning services use among university students, thus helping decision makers plan ahead, prioritize tasks and make better decisions.Originality/valueKnowledge about students' intention to use e-government learning services is even useful for the implementation of other services by higher education institutions. This study also offers some implications for practitioners and researchers concerned about the application of e-government in students' learning development.


2017 ◽  
Vol 10 (7) ◽  
pp. 56 ◽  
Author(s):  
Patrizia Grifoni ◽  
Fernando Ferri ◽  
Tiziana Guzzo

The Internet is deeply changing how buyers and sellers interact in the marketplace. The Web enables consumers to be informed on their purchases both online and offline thanks to crowdsourced reviews. However, recent studies have found evidence that online consumers review could be not truthful as some users such as owners, competitors, paid users, sometimes post fake reviews. In this context the question of credibility is becoming more and more relevant in the Web 2.0 environment in which the concepts of social influence and electronic word of mouth are acquiring a great importance. The user’s perception of online reviews can influence source credibility and the perception of the quality of a product/service, as well as the likelihood that someone will purchase the product/service. This study proposes a model that analyses elements that influence online information credibility and the impact of the perceived credibility on purchase intention.


2019 ◽  
Vol 1 (1) ◽  
pp. 61-72
Author(s):  
Eko Setiawan ◽  
Darius Antoni ◽  
Ahmad Haidar Mirza

Technology Acceptance Model (TAM) dan Webqual merupakan teori penerimaan  teknologi yang digunakan dalam penelitian ini. Terdapat lima konstruk yang  digunakan dan merupakan gabungan dari konstruk-konstruk asli yang terdapat dalam TAM dan Webqual. Kelima konstruk tersebut adalah kemudahan pengguna persepsi (perceived ease of use / PEOU), kegunaan persepsi (perceived  usefulness / PU), kualitas interaksi layanan  (Service Interaction Quality of Website/ Web-SQ), kualitas informasi (Information Quality of Website/ Web-IQ), dan minat perilaku menggunakan teknologi (behavioral intention to use / BIUS). Model gabungan ini selanjutnya digunakan untuk meneliti sistem ujian online berbayar dengan studi kasus qualitiva.id. Penelitian ini selanjutnya akan menjelaskan tentang penerimaan sistem ujian online berbayar  pada pelajar, mahasiswa dan masyarakat umum. Jenis penelitian yang digunakan dalam penelitian ini adalah penelitian eksplanatori dengan teknik analisis data menggunakan pendekatan PLS. Data diperoleh dari responden yang merupakan siswa LKP PalComTech Palembang dengan menyebarkan kuesioner tertutup. Terdapat 142 sampel penelitian yang dianalisis menggunakan metode PLS dengan software smartPLS. Berdasarkan hasil analisis data diperoleh hasil sebagai berikut: (1) Konstruk PEOU mempengaruhi konstruk AT; (2) Konstruk PU mempengaruhi konstruk AT; (3) Konstruk AT pengaruh terhadap konstruk BIUS; (4) Konstruk Web-SQ  tidak memiliki  pengaruh  terhadap  konstruk BIUS;  (5)  Konstruk Web-IQ mempengaruhi konstruk BIUS; dan (6) Konstruk BIUS tidak memiliki pengaruh terhadap konstruk BV;. Selain itu koefisien variabel laten PU terhadap AT memiliki nilai paling besar diantara nilai koefisien variabel laten pada model hubungan antar konstruk lainnya


2021 ◽  
Vol 31 (5) ◽  
pp. 1318
Author(s):  
I Made Fery Aditya ◽  
I Made Pande Dwiana Putra

E-commerce is a general concept in the process of buying and selling products or services through the internet network. The Technology Acceptance Model theory explains the factors that influence the behavior of users of information systems, The purpose of this study is to obtain empirical evidence of the effect of perceived usefulness, perceived ease of use, subjective norms, and quality of information on interest in using e-commerce. This research was conducted on all active college students majoring in Accounting class 2016 at the Faculty of Economics and Business, Udayana University, totaling 115 college students. The sampling method used in this study was saturated sampling and the data analysis technique in this study used multiple linear regression analysis. Based on the analysis conducted, it was found that perceived usefulness, perceived ease of use, subjective norms and quality of information had a positive effect on the interest in using e-commerce. Keywords: Perceived Usefulness; Perceived Ease Of Use; Subjective Norms; Information Quality; E-Commerce.


2019 ◽  
Vol 15 (2) ◽  
pp. 131
Author(s):  
Nera Marinda Machdar

<p><em>The purpose of this study aims to empirically test the effect of information quality on perceived usefulness, the effect of information quality on perceived ease of use, and the effect of perceived ease of use on perceived usefulness. This study uses the Technology Acceptance Model (TAM) developed by Davis (1999) to evaluate perceived usefulness and perceived ease of use.The method of data analysis in this research using analysis of Structural Equation Modeling (SEM) - Partial Least Square (PLS) with PLS 3.0.The resultd of this study showed that (1) Quality of information positively affects perceived usefulness, (2) Quality of information positively affects perceived ease of use, and (3) Perceived ease of use positively affects perceived usefulness. The limitations of this study were as follows: (1) The inherent limitations of the data obtained through the questionnaire, due to differences in the authors' perceptions ofthe research respondents, (2) Limitations of the number of accounting student respondents who are following the course of Accounting Introduction Practicum, and (3) Limitations on the selection of accounting software samples used are limited only to the use of the ACCURATE software. Subsequent research suggested that (1) Consider other independent variables that may influence perceived usefulness and perceived ease of use, such as user satisfaction and user intent, and (2) Increase the sample size of respondents in this case not only students who are taking the course Practicum Introduction to Accounting, but also students who have taken the course before</em></p>


Sign in / Sign up

Export Citation Format

Share Document