In recent years, the expansion of online media has presented firms with rich and voluminous new datasets with profound business applications. Among these, online reviews provide nuanced details on consumers' interactions with products. Analysis of these reviews has enormous potential, but the enormity of the data and the nature of unstructured text make mining these insights challenging and time-consuming. This paper presents three studies examining this problem and suggesting techniques for automated extraction of vital insights. The first study examines the problem of identifying mentions of safety hazards in online reviews. Discussions of hazards may have profound importance for firms and regulators as they seek to protect consumers. However, as most online reviews do not pertain to safety hazards, identifying this small portion of reviews is a challenging problem. Much of the literature in this domain focuses on selecting "smoke terms," or specific words and phrases closely associated with the mentions of safety hazards. We first examine and evaluate prior techniques to identify these reviews, which incorporate substantial human opinion in curating smoke terms and thus vary in their effectiveness. We propose a new automated method that utilizes a heuristic to curate smoke terms, and we find that this method is far more efficient than the human-driven techniques. Finally, we incorporate consumers' star ratings in our analysis, further improving prediction of safety hazard-related discussions. The second study examines the identification of consumer-sourced innovation ideas and opportunities from online reviews. We build upon a widely-accepted attribute mapping framework from the entrepreneurship literature for evaluating and comparing product attributes. We first adapt this framework for use in the analysis of online reviews. Then, we develop analytical techniques based on smoke terms for automated identification of innovation opportunities mentioned in online reviews. These techniques can be used to profile products as to attributes that affect or have the potential to affect their competitive standing. In collaboration with a large countertop appliances manufacturer, we assess and validate the usefulness of these suggestions, tying together the theoretical value of the attribute mapping framework and the practical value of identifying innovation-related discussions in online reviews. The third study addresses safety hazard monitoring for use cases in which a higher yield of safety hazards detected is desirable. We note a trade-off between the efficiency of hazard techniques described in the first study and the depth of such techniques, as a high proportion of identified records refer to true hazards, but several important hazards may be undetected. We suggest several techniques for handling this trade-off, including alternate objective functions for heuristics and fuzzy term matching, which improve the total yield. We examine the efficacy of each of these techniques and contrast their merits with past techniques. Finally, we test the capability of these methods to generalize to online reviews across different product categories. / Doctor of Philosophy / This dissertation presents three studies that utilize text analytic methods to analyze and derive insights from online reviews. The first study aims to detect distinctive words and phrases particularly prevalent in online reviews that describe safety hazards. This study proposes algorithmic and heuristic methods for identifying words and phrases that are especially common in these reviews, allowing for an automated process to prioritize these reviews for practitioners more efficiently. The second study extends these methods for use in detecting mentions of product innovation opportunities in online reviews. We show that these techniques can used to profile products based on attributes that differentiate them from competition or have the potential to do so in the future. Additionally, we validate that product managers find this attribute profiling useful to their innovation processes. Finally, the third study examines automated safety hazard monitoring for situations in which the yield or total number of safety hazards detected is an important consideration in addition to efficiency. We propose a variety of new techniques for handling these situations and contrast them with the techniques used in prior studies. Lastly, we test these methods across diverse product categories.

The purpose of this dissertation intends to discover as well as categorize safety concern reports in online reviews by using key terms prevalent in sub-categories of safety concerns. This dissertation extends the literature of semi-automatic text classification methodology in monitoring and classifying product quality and service concerns. We develop various text classification methods for finding key concerns across a diverse set of product and service categories. Additionally, we generalize our results by testing the performance of our methodologies on online reviews collected from two different data sources (Amazon product reviews and Facebook hospital service reviews). Stakeholders such as product designers and safety regulators can use the semi-automatic classification procedure to subcategorize safety concerns by injury type and narrative type (Chapter 1). We enhance the text classification approach by proposing a Risk Assessment Model for quality management (QM) professionals, safety regulators, and product designers to allow them to estimate overall risk level of specific products by analyzing consumer-generated content in online reviews (Chapter 2). Monitoring and prioritizing the hazard risk levels of products will help the stakeholders to make appropriate actions on mitigating the risk of product safety. Lastly, the text classification approach discovers and ranks aspects of services that predict overall user satisfaction (Chapter 3). The key service terms are beneficial for healthcare providers to rapidly trace specific service concerns for improving the hospital services. / Doctor of Philosophy / This dissertation extends past studies by examining safety surveillance of online reviews. We examine online reviews reporting specific categories of safety concerns and contrast them with reviews not reporting these specific safety concerns. Business and regulators are benefited in detecting, categorizing, and prioritizing safety concerns across product categories. We use key terms prevalent in domain-related safety concerns for granular analysis of consumer reviews. Secondly, beyond utilizing the key terms to discover specific hazard incidents, safety regulators and manufacturers may use the extended risk assessment framework to estimate the risk severity, risk likelihood, and overall risk level of a specific product. The model could be useful for product safety practitioners in product risk identification and mitigation. Finally, this dissertation identifies the aspects of service quality concerns present in online hospital reviews. This study uses text analytics method by using key terms to detect these specific service concerns and hence determine primary rationales for patient feedback on hospital services. Managerially, this information helps to prioritize the areas in greatest need of improvement of hospital services. Additionally, generating key terms for a particular service attribute aids health care policy makers and providers in rapidly monitoring specific concerns and adjusting policies or resources to better serve patient

Yes / Product reviews are posted online by the hundreds and thousands for popular products. Handling such a large volume of continuously generated online content is a challenging task for buyers, sellers and researchers. The purpose of this study is to rank the overwhelming number of reviews using their predicted helpfulness scores. The helpfulness score is predicted using features extracted from review text, product description, and customer question-answer data of a product using the random-forest classifier and gradient boosting regressor. The system classifies reviews into low or high quality with the random-forest classifier. The helpfulness scores of the high-quality reviews are only predicted using the gradient boosting regressor. The helpfulness scores of the low-quality reviews are not calculated because they are never going to be in the top k reviews. They are just added at the end of the review list to the review-listing website. The proposed system provides fair review placement on review listing pages and makes all high-quality reviews visible to customers on the top. The experimental results on data from two popular Indian e-commerce websites validate our claim, as 3–4 newer high-quality reviews are placed in the top ten reviews along with 5–6 older reviews based on review helpfulness. Our findings indicate that inclusion of features from product description data and customer question-answer data improves the prediction accuracy of the helpfulness score. / Ministry of Electronics and Information Technology (MeitY), Government of India for financial support during research work through “Visvesvaraya PhD Scheme for Electronics and IT”.

Dans un contexte d’achat sur internet, la communication du bouche-à-oreille électronique dans sa forme la plus usuelle, c’est-à-dire par les avis de consommateurs en ligne, constitue une source d’information importante sur les biens et les services. Les avis en ligne possèdent un fort potentiel d’influence sur les choix, les évaluations et la prise de décision des consommateurs. Cependant, dans un environnement virtuel caractérisé par l’anonymat des sources et l’existence de manipulations frauduleuses avérées de la part de certaines entreprises, les internautes remettent en question le bien-fondé de cette source d’information. Par ailleurs, hormis la question cruciale de la crédibilité que l’on peut octroyer aux avis des consommateurs sur internet, la recherche dans ce domaine continue à s’interroger sur les mécanismes d’influence des avis d’anciens acheteurs ou utilisateurs du produit sur le comportement du consommateur. Bien qu’un nombre important de travaux ait été dédié à ce sujet, les résultats de ces études continuent d'attiser la curiosité des chercheurs quant aux effets des avis des consommateurs sur les réactions des internautes que l'on croyait globalement avoir identifiés. Plusieurs résultats contradictoires ont en effet été observés dans ce champ de recherche. C’est la raison pour laquelle, à partir des limites observées dans les travaux antérieurs et les orientations préconisées par d’autres, cette recherche se propose d’étudier sous une approche expérimentale l’influence de certaines caractéristiques des avis en ligne sur le comportement du consommateur. Le but est d’établir diverses conditions d’exposition aux avis de consommateurs pouvant conduire à une meilleure compréhension de leur impact sur les perceptions et réactions des internautes. Globalement, nos résultats auprès d’un échantillon de 503 répondants actifs permettent de valider les hypothèses de la recherche avancée et offrent une analyse intéressante des effets de la perception des avis en ligne sur les réactions et les intentions comportementales des internautes. / In an online shopping context, the word-of-mouth in its most usual form, ie online consumer reviews, is an important source of information on goods and services. Online reviews have a strong potential to influence consumers' choices, evaluations and decision-making. However, in a virtual environment marked by the anonymity of sources and the existence of fraudulent manipulations proved by some companies, Internet users question the validity of this source of information. Moreover, apart from the crucial question of the credibility that can be given to the opinions of other consumers on the Internet, research in this field continues to examine the influence mechanisms of former buyers or product users’ opinions on consumer behavior. A considerable amount of research has been dedicated to this subject. The results of these studies continue to stir researchers' curiosity about the effects of consumer opinions on the reactions of Internet users who were generally believed to have identified. Several contradictory results have been observed in this field of research. This is why, based on the limits observed in previous researches and orientations recommended by others, this thesis proposes to study under an experimental approach the influence of some online consumers’ characteristics. The aim is to create different exposure conditions to consumer opinions that can lead to a better understanding of the influence they can have on the perceptions and reactions of Internet users. Overall, our findings from a sample of 503 active respondents highlight this advanced hypothesis and provide an interesting analysis of the effects of e-reviews perceptions on Internet users’ responses and behavioral intentions.

Previous research has established that punctuation can be used to communicate nuances of meaning in online writing (McAndrew & De Jonge, 2011). Punctuation, considered a computer mediated communication (CMC) cue, expresses tone and emotion and disambiguates an author's intention (Vandergriff, 2013). Quotation marks as CMC cues can serve pragmatic functions and have been understudied. Some of these functions have been generally described (Predelli, 2003). However, no corpus study has specifically focused on the pragmatic uses of quotations in online text. Consumer reviews, a genre of online text, can directly impact business profits and influence customers' purchasing decisions (Floyd, Freling, Alhoqail, Cho & Freling, 2014). Businesses are investing in sentiment analysis to gauge their target market's opinions (Salehan & Kim, 2016). Sentiment analysis is the computerized appraisal of a text to determine whether its author is expressing a positive or negative opinion (Novak, Smailovic, Sluban & Mozetic, 2015). Sentiment analysis programs are still limited and could be improved in accuracy. Most programs rely on lexicons of words given a pre-determined polarity value (positive or negative) out of context (Novak et al., 2015). However, context is crucial to communication, and sentiment analysis programs could incorporate a better variety of contextual linguistic features to improve their accuracy. Quotations used for pragmatic communication is such a feature. This study discovered seven pragmatic quotation uses in a 2014 Yelp review corpus: Collective Knowledge, Non-standard, Grammatical, Non-literal, Narrative, Idiolect, and Emphasis. An ANOVA and Tukey HSD test were performed, and the results were significant. Pragmatic category accounted for 15% of the variance in review star rating. The Collective Knowledge category and the Narrative and Non-literal categories were significantly different from each other. The Collective Knowledge category showed a correlation with positive sentiment, while the Narrative and Non-literal categories displayed a correlation with negative sentiment. These three categories are likely present in several types of online text, making them valuable for further sentiment analysis research. If these pragmatic patterns could be detected automatically, they could be used in sentiment algorithms to give a more accurate picture of author opinion.

This thesis aims to look closer upon how hotels manage electronic Word of Mouth and suggests potential strategies. Focus is being put on online guest reviews and the researchintends to study the impact of online reviews in the hospitality industry as well as defining the possibilities and challenges. The aim is not to make generalizations for the entire industry, but to be a suggestion for how to manage online reviews. Case studies have been performed withsix different hotels with different characteristics, to be able to answer the research questions. In order to provide an understanding of the topic, theories concerning e-Word of Mouth marketing, has been chosen. The traditional Word of Mouth and electronic Word of Mouth isdefined and the possibilities/challenges are explained. Theories on management of online reviews are also presented in the thesis. The empirical research has shown that e-Word of Mouth is still a new phenomenon in the hospitality industry. However most respondents agree on the influence of online recommendations which make it important to monitor and track them. The main possibility of managing online reviews is the chance of increased revenues and the ability to create brand awareness. It helps hotels to understand the needs of their customers. However challenges of sincerity and the big amount of information that is needed to control, are problematic. The research has shown that only few respondents work actively with some type of e-Word of Mouth. These results have provided the thesis with suggestions on management of online reviews.

Background: To understand the consumers’ motivation to write online reviews is of importance, especially for companies since a large number of reviews have a positive influence on sales. Previous research has been done regarding what motivate consumers to provide user generated content, online word of mouth and also, to some extent, online reviews. However, these studies have primarily been adopted in a quantitative manner. To explore, from customers’ own perspective, the motivation to write online reviews is therefore valuable to add depth to the existing literature. Purpose: The purpose of this study is to explore customers’ motivation to write online reviews. Research question: What factors motivate customers to write online reviews? Methodology: The design of the research is a case study where the data collection method was conducted by semi-structured interviews. Conclusion: The result of this study shows that customers’ motivation to write online reviews is due to a variety of situations. The customers are motivated to write to enhance their selfimage, helping both customers and companies, and in some situations to even harm companies. Also, customers were motivated to write to obtain economical incentives. The features of the platform are important, where easiness and the opportunity to be anonymous were preferred.

Due to the popularity of social media and an increase in the engagement of social care, traditional word-of-mouth communications has been replaced by electronic word-of-mouth (e-WOM). Facebook, the most popular website in the United States, is home to nearly 18 million brand or business pages and may be accessed by social media-users aiming to engage in social care, which is customer service via social media. Extending existing research, this study employed in-depth interviews to determine whether or not social media-users are affected by the feedback of other users on restaurants' Facebook pages. The results of this study suggest that Facebook is being used as a tool to attain user feedback regarding restaurants and is perceived as a credible tool. The results also suggest that social media-users are mainly affected by others' user feedback when they are researching a restaurant they have not yet experienced. Finally, the findings of this study suggest that restaurants using Facebook should respond to all types of user feedback, since this practice may result in providing social media-users with a more positive perception of the restaurant.

Konsumenter använder och förlitar sig i allt större utsträckning på andras åsikter som lagtsupp på recensionssidor på nätet. Nyligen har dock skandaler uppmärksammat förekomsten avmanipulation av recensioner vilket lett till att recensionssajters trovärdighet kan ifrågasättas.Tidigare forskning har också gett tvetydiga svar på huruvida folk litar på recensioner ellerinte.Att utforska det här ämnet är viktigt eftersom marknadsförare måste förstå sig påkonsumenters förtroende för onlinerecensioner då dessa kan användas som ett kraftfulltmarknadsföringsverktyg och som ett nytt element i marknadsföringsmixen. Den här studiensyftar till att utforska konsumenters förtroende för onlinerecensioner genom att undersökavarför man litar på recensioner och vad som gör en onlinerecension pålitlig. Genomanvändandet av en modifierad version av Shannon och Weavers kommunikationsmodell, kanen recension struktureras kring elementen meddelande, sändare, mottagare, kanal ochåterkoppling. Den här studien implementerar en kvalitativ metod där semistruktureradeintervjuer använts för att få en djupare förståelse för intervjuobjektens känslor och tankar.Studiens resultat indikerar att konsumenter inte litar på individuella recensioner, utan snararepå systemet med onlinerecensioner i sin helhet; antalet recensioner nämndes som den mestinflytelserika faktorn för att skapa tillit. Studien visar också att den som läser recensionerframförallt använder meddelandet för att avgöra om en recension är pålitlig eller inte. Pågrund av avsaknaden av information om personen som skriver recensionen är meddelandetdet enda läsaren verkligen kan utvärdera. / Increasingly, consumers are using and relying on the opinions of others posted on onlinereview sites. However, recent scandals has brought attention to the existence of reviewmanipulation and questioned the credibility of online review sites. Furthermore, previousresearch has shown conflicting findings on whether or not consumers trust online reviews.Exploring these matters becomes important since marketers need to be able to understandconsumer trust in online reviews as it then can be used as a powerful marketing tool and as anew element of the marketing communication mix. The aim of this study is to exploreconsumer trust in online reviews by investigating why people trust online reviews, and whatmakes an online review trustworthy. This is accomplished using a modified version of theShannon and Weaver communication model which is structured around the elementsmessage, sender, receiver, channel and feedback. The study employs a qualitative methodusing semi-structured interviews in order to gain a deeper understanding of the thoughts andfeelings of the interviewees. The findings indicate that consumers do not put their trust inindividual reviews, but in the online review system as a whole; the number of reviews postedwas mentioned as the most influential factor for creating trust. The study shows that readersof reviews primarily use the message to determine whether a review is trustworthy or not.Because of the lack of information about the person writing the online review, it is the onlything the reader can truly evaluate.

The purpose of this study was to provide a better understanding why consumers conduct research online and post online reviews. This was achieved by finding relationships between the two phenomena presented above. The study used a cross-sectional design since little was known about the relationships within the chosen research area. Results were measured by quantitative instrument through a self-administered questionnaire divided by each phenomenon. The questionnaire was directed towards Swedish consumers in the buying decision process.Consumers conduct research online with the specific goal for gaining information posted by other consumers, since it has a higher credibility. Helping other consumers and self-expression was reasons found in relation to why consumers post online reviews.Findings strengthen the fact that higher education could be found of relationship to conduct research online; however, not to post online reviews. In comparison to previous studies the study measured online shopping frequency instead of Internet experience. It was found that consumers who conducted online shopping were in relation with previous presented theories. / Syftet med denna studie var att skapa en bättre förståelse till varför konsumenter publicerar och konsumerar onlinerecensioner. Genom att applicera en tvärsnittsdesign kunde syftet uppnås då relationer identifierades. För att möjliggöra mätning av resultatet applicerades ett kvantitativt instrument genom en självadministrerad enkät vilken var uppdelad efter respektive fenomen.Onlinerecensioner används för det specifika målet att samla större mängder information innan köpbeslut. De onlinerecensioner som används främst är de skapade av andra konsumenter då de anses vara av högre trovärdighet än information skapat av företag. Att hjälpa andra konsumenter och självförbättring är vad resultatet påvisar vara anledning till att konsumenter publicerar onlinerecensioner.Vidare stärkte resultatet en relation där konsumenter med högre utbildning tenderar att konsumera onlinerecensioner men inte till att publicera dessa. Konsumenternas köpfrekvens applicerades istället för att mäta graden erfarenhet av Internet, vilket gjorts i tidigare studier.