No description available.

Review helpfulness prediction has attracted growing attention of researchers that proposed various solutions using Machine Learning (ML) techniques. Most of the studies used online reviews from Amazon to predict helpfulness where each review is accompanied with information indicating how many people found the review helpful. This research aims to analyze the complete process of modelling review helpfulness from several perspectives. Experiments are conducted comparing different methods for representing the review text as well as analyzing the importance of data sampling for regression compared to using non-sampled datasets. Additionally, a set of review, review meta-data and product features are evaluated on their ability to capture the helpfulness of reviews. Two Amazon product review datasets are utilized for the experiments and two of the most widely used machine-learning algorithms, Linear Regression and Convolutional Neural Network (CNN). The experiments empirically demonstrate that the choice of representation of the textual data has an impact on performance with tf-idf and word2Vec obtaining the lowest Mean Squared Error (MSE) values. The importance of data sampling is also evident from the experiments as the imbalanced ratios in the unsampled dataset negatively affected the performance of both models with bias predictions in favor of the majority group of high ratios in the dataset. Lastly, the findings suggest that review features such as unigrams of review text and title, length of review text in words, polarity of title along with rating as review meta-data feature are the most influential features for determining helpfulness of reviews.

In this study we gather customer reviews from Prisjakt, a Swedish price comparison site, with the goal to study the relationship between review and rating, known as sentiment analysis. The purpose of the study is to evaluate three different supervised machine learning models on a fine-grained dependent variable representing the review rating. For classification, a binary and multinomial model is used with the one-versus-one strategy implemented in the Support Vector Machine, with a linear kernel, evaluated with F1, accuracy, precision and recall scores. We use Support Vector Regression by approximating the fine-grained variable as continuous, evaluated using MSE. Furthermore, three models are evaluated on a balanced and unbalanced dataset in order to investigate the effects of class imbalance. The results show that the SVR performs better on unbalanced fine-grained data, with the best fine-grained model reaching a MSE 4.12, compared to the balanced SVR (6.84). The binary SVM model reaches an accuracy of 86.37% and weighted F1 macro of 86.36% on the unbalanced data, while the balanced binary SVM model reaches approximately 80% for both measures. The multinomial model shows the worst performance due to the inability to handle class imbalance, despite the implementation of class weights. Furthermore, results from feature engineering shows that SVR benefits marginally from certain regex conversions, and tf-idf weighting shows better performance on the balanced sets compared to the unbalanced sets.

The Internet is becoming the main source for various tasks, from learning, to working, and shopping. There are many websites one can use to shop. Almost all stores have a website from which you can order anything you might want. As online shopping becomes more prominent, it is important to understand the effects of the Internet and its product reviewers and, specific to this study, consumer decision making. This study seeks to understand the effect of star ratings and valence on review perception between the different cognitive levels of individuals. Recognition review perception, and intent to purchase were being measured. Results showed that need for cognition did have an effect on accuracy of recognition and perceived valence. Need for cognition and congruency as well as actual valence had an effect on perceived valence. Need for cognition, actual valence, and congruency all had an effect on purchase intention. This research is important because it is relevant to a growing trend around the world. Technology is already integrated into nearly everyone’s lives and it is only going to more so as we continue to evolve. Just as it is becoming more common for people to receive education from online institutions, and for employers to use more Internet based applications, it is only natural consumers will continue the trend of purchasing items online. Learning the social and cognitive influences of online reviews on perception and purchasing intentions is something everyone needs to be aware of.

Online word-of-mouth (WOM) platforms have been referred to by various terms such as online communities, feedback systems, peer reputation systems, or consumer generated media. Such systems provide a global platform for customers to share their experiences, and also rate service providers. WOM systems are burgeoning on the Internet for products such as music and books (Amazon.com), news (Slashdot.org), consumer electronics (shopping.com), tourism and travel (Tripadvisor.com; Hotels.com), and many other products and services. As with the traditional WOM, numerous studies have shown that these systems have a significant impact on customer decision making process, their satisfaction with goods and services, and the overall value of online economic transactions. In this study, the primary focus were the product review systems (PRS). These review systems are less personal but more ubiquitous platforms for online WOM wherein consumers post reviews about the products/services they have consumed. These reviews are widely accessible to other consumers but are disseminated only when other consumer consult these reviews during the purchasing process. However, there are still numerous problems associated with these systems. Recent studies have shown that there are numerous instances of deceptive information provided by service providers themselves or customers who have been paid by commercial parties. Added to this is the problem of anonymity in a computer mediated environment that adds to the already existing uncertainty for the consumer. Further, each review system consists of hundreds of consumer reviews associated with any given product or service. Given that consumers face these numerous problems, research is yet to examine the factors that drive the consumers develop trust in these reviews, and base their purchasing decisions on the information gleaned from the review systems. The main objective of this study was to explore this interesting phenomenon. To this end, this study applied uncertainty reduction theory and Social identity theory to delineate certain aspects of the online reviews that might have an impact on the consumer's assessment of online product reviews. Based on these theories, it was hypothesized that the informational content of the review and social component of the review (individuals' identity information disclosure and the consumers' perceived similarity with this information) have a significant effect on the consumers' trust in a review and subsequently the purchase intention. Further, based on the elaboration likelihood model, it was also posited that consumers' use of these heuristics is more salient while evaluating high involvement products than low involvement products. To test the hypotheses, the study adopted a quasi-experimental design with 2x2 (2 levels each for information content and social component within-subjects) x 2 (2 involvement modes between-subjects) full factorial design. Based on two levels for each of these factors, four reviews similar to those found in sites such as tripadvisor.com were created. A total of 283 students (153 in high involvement mode and 130 in low involvement mode) evaluated these reviews and assigned trust scores as well purchase intention scores to each review. The data was analyzed using linear mixed models and structural equation modeling. The results showed that both the main effects, information content of the review, and the consumers' perceived social identity with the reviewer contribute to an increased trust in the reviews. The study data did not support the hypothesis that involvement of the activity moderates the above mentioned relationships. Within this, information content was found to be playing an important role in both the involvement modes whereas the social component explained more variance in the trust in the high involvement mode than low involvement mode. Some of the results concur with previous research in both traditional and online WOM. The significance of these results in the extant literature as well their implications for both product review system providers as well tourism and hospitality service providers are discussed in detail. / Business Administration

Numerical information is ubiquitous in modern life. The prevalence of numerical information in the marketplace necessitates understanding how consumers handle and interpret that information, for both theoretical and practical reasons. Past research has largely focused on consumers’ encoding of numbers, calculative limitations, and usage of heuristics. This dissertation will contribute to this burgeoning literature in several ways. First, I identify a general tendency in how consumers calculate ratios based on an intuitive model of division. Specifically, consumers tend to divide larger numbers by smaller numbers. The intuitive model of division has marketing implications for both consumers’ evaluations of quantity offers and sensitivities to promotions. Next, I examine how consumers draw inferences from distributional information. In contrast to the assumption that consumers utilize means to assess central tendency, I demonstrate that consumers use the modal response to judge what is typical, with implications for consumers’ inferences about product ratings and other social distributions. / PHD / Numerical information is ubiquitous in modern life. The prevalence of numerical information in the marketplace necessitates understanding how consumers handle and interpret that information, for both theoretical and practical reasons. Past research has largely focused on how consumers’ mentally perceive numbers, how difficult it is to engage in calculation, and usage of mental shortcuts. This dissertation will contribute to this burgeoning literature in several ways. First, I identify a general tendency in how consumers calculate ratios based on an intuitive model of division. Specifically, consumers tend to divide larger numbers by smaller numbers. The intuitive model of division has marketing implications for both consumers’ evaluations of quantity offers and sensitivities to promotions. Next, I examine how consumers draw inferences from distributional information. In contrast to the assumption that consumers utilize means to assess central tendency, I demonstrate that consumers use the modal response to judge what is typical, with implications for consumers’ inferences about product ratings and other social distributions.

No description available.