Expressing Sentiments in Game Reviews

Secui, Ana; Sirbu, Maria-Dorinela; Dascălu, Mihai; Crossley, Scott A.; Ruşeţi, Ştefan; Trăușan-Matu, Ștefan

doi:10.1007/978-3-319-44748-3_35

Cited by 6 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ruseti et al [10] used traditional support vector machine, multinomial Naive-Bayes, and deep neural networks (DNN) for triple classification of pre-processed game review texts. Secui et al [11] analyzed 9750 reviews using their own constructed model. Eight affective components were identified using a Principal Component Analysis (PCA) and a Discriminant Function Analysis based on the emerging components classified game reviews into three categories with a 55% accuracy.…”

Section: Sentiment Analysis Of Game Reviewsmentioning

confidence: 99%

Sentiment analysis of Chinese game reviews based on SKEP

Wang¹

2023

AETR

View full text Add to dashboard Cite

As China's game industry enters the era of the stock market, improving the quality of game products in all aspects is gradually becoming an important part of the development of China's game industry. There exists a large amount of game review data on the Internet, and sentiment analysis of them can provide important help for game makers to make decisions. However, the huge scale of game reviews and the extensive use of metaphors and sarcasm make it difficult to conduct sentiment analysis. In this paper, we experiment with a publicly available Chinese game review dataset using the ERNIE pre-training model equipped with the SKEP method, which improves the model's masking strategies and pre-training objectives to specialize the model's sentiment analysis ability. In addition, we compared the performance of ERNIE1.0, RoBERTa, MacBERT, BERT-wwm, and Nezha. The precision, recall, and F1 score of SKEP are higher than those of the models for comparison studies, which confirms the effectiveness of SKEP in this area.

show abstract

Section: Sentiment Analysis Of Game Reviewsmentioning

confidence: 99%

Sentiment analysis of Chinese game reviews based on SKEP

Wang¹

2023

AETR

View full text Add to dashboard Cite

show abstract

“…For the experiments, we used a game reviews dataset containing context textual data posted on the MetaCritic website (https://www.metacritic.com/, accessed on 28 September 2021). The original version of this dataset is presented in [39] and improved in [40]. From the dataset, we use only the reviews and the polarity assigned for each review, although the collected raw data also contain other information.…”

Section: Datasetmentioning

confidence: 99%

Topic-Based Document-Level Sentiment Analysis Using Contextual Cues

et al. 2021

View full text Add to dashboard Cite

Document-level Sentiment Analysis is a complex task that implies the analysis of large textual content that can incorporate multiple contradictory polarities at the phrase and word levels. Most of the current approaches either represent textual data using pre-trained word embeddings without considering the local context that can be extracted from the dataset, or they detect the overall topic polarity without considering both the local and global context. In this paper, we propose a novel document-topic embedding model, DocTopic2Vec, for document-level polarity detection in large texts by employing general and specific contextual cues obtained through the use of document embeddings (Doc2Vec) and Topic Modeling. In our approach, (1) we use a large dataset with game reviews to create different word embeddings by applying Word2Vec, FastText, and GloVe, (2) we create Doc2Vecs enriched with the local context given by the word embeddings for each review, (3) we construct topic embeddings Topic2Vec using three Topic Modeling algorithms, i.e., LDA, NMF, and LSI, to enhance the global context of the Sentiment Analysis task, (4) for each document and its dominant topic, we build the new DocTopic2Vec by concatenating the Doc2Vec with the Topic2Vec created with the same word embedding. We also design six new Convolutional-based (Bidirectional) Recurrent Deep Neural Network Architectures that show promising results for this task. The proposed DocTopic2Vecs are used to benchmark multiple Machine and Deep Learning models, i.e., a Logistic Regression model, used as a baseline, and 18 Deep Neural Networks Architectures. The experimental results show that the new embedding and the new Deep Neural Network Architectures achieve better results than the baseline, i.e., Logistic Regression and Doc2Vec.

show abstract

“…In both experimental and observational research, the linguistic tags used online correlate with actions such as levels of adult playfulness (Proyer & Brauer, 2018), the stress level of academics based on their publications (Ratsamy et al, 2018), and learning outcomes in online courses (Lee & Recker, 2018). Focused on game communities, however, psycholinguistic and linguistic markers can also be used to estimate player enjoyment during gameplay (Secui et al, 2016).…”

Section: Communities Within Social Mediamentioning

confidence: 99%

“…Linguistic markers have also been used in research to identify participants with negative dysfunctional dispositions on Facebook (Akhtar et al, 2018), the growth of racist intent on forums (Bäck et al, 2018), or even the behavior of Twitter users with influenza (Flekova et al, 2018). In the domain of social media, his approach can also be used to estimate whether or not a player enjoyed playing a game (Secui et al, 2016). In gaming, the majority of the research in this area is retrospective and correlative.…”

Section: Introductionmentioning

confidence: 99%

“Lock, Load, n’ Thank the Driver”: The Positive Influence of Prosocial Activity on Language in Online Social Groups

Ryan

McDonnell

Meade

2020

Social Media + Society

View full text Add to dashboard Cite

Social-psychological research aims to understand and improve human functioning in a wide variety of contexts. With research showing that trust, joy, and anticipation are among the most influential positive emotions in aspects of online communication, determining the environmental factors that may influence communication and interaction is of significant interest. This research aims to identify the ways prosocial actions influence language within an online social community. Using a longitudinal observational analysis of linguistic markers, an analysis was done on an online community, surrounding the game “Fortnite.” At different time intervals, the written dialogue was analyzed for psycholinguistic markers through LIWC to detect variance in tone within the community. Users were queried ( n = 9,037), filtered to meet the inclusion criteria (n = 7,221) and having their posts (n = 1,232,741) assigned to three cohorts, those of pre-prosocial, post-prosocial, and post-reciprocity, depending on the posting date. A statistically significant ( R2 = .572) increase in positive emotion linguistic markers was detected in the data set after the prosocial action was added. A change in motivation did not reduce the use of positive emotion linguistic markers, contrary to what was expected. This research observes how individuals within online communities treat each other and that interaction can be influenced to improve the participant’s experience and the quality of dialogue. This research addresses how game design and human–computer interaction influence human communication and behavior. It supports that small changes to the real world have ripple effects in online communities for the better.

show abstract

Expressing Sentiments in Game Reviews

Cited by 6 publications

References 7 publications

Sentiment analysis of Chinese game reviews based on SKEP

Sentiment analysis of Chinese game reviews based on SKEP

Topic-Based Document-Level Sentiment Analysis Using Contextual Cues

“Lock, Load, n’ Thank the Driver”: The Positive Influence of Prosocial Activity on Language in Online Social Groups

Contact Info

Product

Resources

About