Movie Rating Research Articles

The Investigation of Progress and Application in the Multi-Armed Bandit Algorithm

This article delves deeply into the development process and practical applications of the multi-armed bandit algorithm in the current digital era. With the continuous popularity of online advertising and online learning, information has grown explosively, making decision optimization crucial. The multi-armed bandit algorithm, as a sequential decision-making model, encompasses common algorithms such as the greedy algorithm, -greedy algorithm, UCB algorithm, and Thompson sampling. Its main role is to seek the best balance between exploration and exploitation to solve the fundamental problems in reinforcement learning. The article introduces an internationally released datasets, namely MovieLens, and elaborates in detail a series of indicators for evaluation, including the average number of friends per user, the average number of listened-to artists per user, the average number of movie rating times, the average number of tags added by users, content diversity indicators, and statistics on the differences in click-through rates of recommendations for different types of movies. In addition, the article also presents the specific methods of literature collection, screening, analysis, and review. Its purpose is to understand the multi-armed bandit algorithm more deeply and provide strong guidance for the future development and wide application of this algorithm in various fields.

On the Performance of the Minimax Optimal Strategy in the Stochastic Case of Logistic Bandits

The multi-armed bandit problem is a well-established model for examining the exploration/exploitation trade-offs in sequential decision-making tasks. This study focuses on the logistic bandit, where rewards are derived from two distinct datasets of movie ratings, ranging from 1 to 5, each characterized by different variances. Previous research has shown that regret bounds for multi-armed bandit algorithms can be unstable across varying environments. However, this paper provides new insights by demonstrating the robustness of the Minimax Optimal Strategy in the Stochastic (MOSS) algorithm across environments with differing reward variances. Unlike prior studies, this research shows that MOSS maintains superior performance in both dense and sparse reward settings, consistently outperforming widely used algorithms like UCB and TS, particularly in high variance conditions and over sufficient trials. The findings indicate that MOSS achieves logarithmic expected regret for both types of environments, effectively balancing exploration and exploitation. Specifically, with K arms and T time steps, the regret R(T) of MOSS is bounded by O((KT logT )). This work highlights MOSS as a robust solution for handling diverse stochastic conditions, filling a crucial gap in the understanding of its practical adaptability across different reward distributions.

Stereotypes on the Small Screen: Applying a Media Rating Tool to Portrayals of Autism in Television Series 2010 – 2019

Images of disability in media, both accurate and stereotypical, have a powerful effect on the opinions and actions of individuals. Although researchers have noted and categorized ways that disability has been portrayed in movies and on television over many decades, little research has focused on more recent media. Furthermore, extant research has focused on physical, sensory, and psychiatric disabilities, leaving autism spectrum disorder (ASD) under-researched. This paper describes and analyzes portrayals of ASD in the past decade (2010–2019) in television series produced primarily in the United States and marketed for popular audiences. This media content analysis utilized a movie rating tool designed to assess the presence or absence of previously researched disability stereotypes (Safran, 2000). Using this tool, trained undergraduate students rated twelve television series spanning eight genres. Numerical and narrative data were used to categorize series as more positive/realistic or more negative/stereotypical. Results revealed ten of the twelve to be overall positive/realistic, although stereotypes of disability persisted throughout all series. ASD is increasingly visible on the small screen; it is imperative that images in media consumed by the general public present disability, and specifically ASD, in accurate, positive, realistic, and non-stereotyped ways. The paper concludes with recommendations toward this goal.

Shifting evaluations: The impact of status increase on quality evaluations in the movie industry

Contrary to the belief that higher status automatically improves quality evaluations, the purpose of this study was to investigate the effects of status on perceived quality in the movie industry. The research hypothesizes that Academy Award-winning films receive less favorable evaluations post-award and that increased popularity post-award exacerbates this effect compared to their nominated-only counterparts. By analyzing over 47,000 reviews from IMDB, covering both winning and nominated-only movies from 2008 to 2019, the study finds that ratings of award-winning movies significantly decline post-award, while those of nominated-only movies do not experience a similar decline. Despite the decline in ratings, sentiment scores in reviews for award-winning movies do not show a significant drop post-award. Additionally, the study observes that the increase in review volume post-award correlates with lower ratings for winning movies, underscoring the impact of varied audience tastes on evaluations. This phenomenon highlights the divergence in evaluation criteria between experts and the general public. The findings emphasize the complexity of status dynamics and suggest that higher status, as conferred by awards, does not necessarily translate to sustained positive evaluations. This research underscores the need for further exploration into the dynamics of status, popularity, and evaluation, particularly in fields where public perception plays a crucial role. Such insights are valuable for understanding how public recognition and increased visibility can influence the longevity of perceived quality in cultural products.

Hybrid recommendation system for movies using artificial neural network

RS are becoming more crucial in everyone’s life as the need for recommendations that reflect users’ interests grows every day. For new users, the movie RS based on a film’s ratings has become an exciting trait. However, scalability, cold start, and sparsity issues are the barriers for RS that must be overcome by proposed hybrid system. The MF Technique was utilizedin the modified SVD to minimizethe numberof features & spatial dimensions in the set of data. To detect the similarity between the movies the Content Driven KNN is proposed based on cosine similarity. To measure the closest neighbor, IKSOM is proposed by using EISEN cosine correlation distance which reduce the cluster overlapping. The K means ++ Clustering is proposed for categorize the data in the movies by silhouette method which calculate the optimal quantity of clusters. The overall hybrid method generates the scores also SVD-CF predicts movie ratings and provides the top recommendation for the movies. The results reveal the proposed hybrid system attains high accuracy, precision, Recall and F1-score also achieves the less error as RMSE and MAE when compared to the existing system.

Analyzing the strengths and weaknesses of diverse algorithms for solving Multi-Armed Bandit problems using Python

With the rapid advancement of science and technology, the internet has become an integral part of daily life, revolutionizing how people access information and make decisions. In this context, algorithms play a pivotal role in helping individuals make informed choices tailored to their preferences across various domains. Utilizing the MovieLens dataset (https://grouplens.org/datasets/movielens/1m/), which contains a rich compilation of movie ratings and metadata, this study conducts a thorough analysis using Python to assess the performance of four distinct algorithms: Explore-then-Commit (ETC), Upper Confidence Bound (UCB), Thompson Sampling (TS), and Epsilon-Greedy. The comparison reveals that the ETC algorithm excels in applications such as online advertising recommendation and autonomous driving. The UCB algorithm proves more advantageous in financial analysis, where risk management is critical. The TS algorithm is particularly effective in short video recommendation systems, while the Epsilon-Greedy algorithm is well-suited for balancing exploration with reward. Overall, the results indicate that the TS algorithm outperforms the others in general efficacy.

Cross-domain information fusion and personalized recommendation in artificial intelligence recommendation system based on mathematical matrix decomposition

Given the challenges of inter-domain information fusion and data sparsity in collaborative filtering algorithms, this paper proposes a cross-domain information fusion matrix decomposition algorithm to enhance the accuracy of personalized recommendations in artificial intelligence recommendation systems. The study begins by collecting Douban movie rating data and social network information. To ensure data integrity, Levenshtein distance detection is employed to remove duplicate scores, while natural language processing technology is utilized to extract keywords and topic information from social texts. Additionally, graph convolutional networks are utilized to convert user relationships into feature vectors, and a unique thermal coding method is used to convert discrete user and movie information into binary matrices. To prevent overfitting, the Ridge regularization method is introduced to gradually optimize potential feature vectors. Weighted average and feature connection techniques are then applied to integrate features from different fields. Moreover, the paper combines the item-based collaborative filtering algorithm with merged user characteristics to generate personalized recommendation lists.In the experimental stage, the paper conducts cross-domain information fusion optimization on four mainstream mathematical matrix decomposition algorithms: alternating least squares method, non-negative matrix decomposition, singular value decomposition, and latent factor model (LFM). It compares these algorithms with the non-fused approach. The results indicate a significant improvement in score accuracy, with mean absolute error and root mean squared error reduced by 12.8% and 13.2% respectively across the four algorithms. Additionally, when k = 10, the average F1 score reaches 0.97, and the ranking accuracy coverage of the LFM algorithm increases by 54.2%. Overall, the mathematical matrix decomposition algorithm combined with cross-domain information fusion demonstrates clear advantages in accuracy, prediction performance, recommendation diversity, and ranking quality, and improves the accuracy and diversity of the recommendation system. By effectively addressing collaborative filtering challenges through the integration of diverse techniques, it significantly surpasses traditional models in recommendation accuracy and variety.

Comparative analysis of the KL-UCB and UCB algorithms: Delving into complexity and performance

This paper embarks on a meticulous comparative exploration of two venerable algorithms often invoked in multi-armed bandit problems: the Kullback-Leibler Upper Confidence Bound (KL-UCB) and the generic Upper Confidence Bound (UCB) algorithms. Initially, a comprehensive discourse is presented, elucidating the definition, evolution, and real-world applications of both algorithms. The crux of the study then shifts to a side-by-side comparison, weighing the regret performance and time complexities when applied to a quintessential movie rating dataset. In the trenches of practical implementations, addressing multi-armed bandit problems invariably demands extensive training. Consequently, even seemingly minor variations in algorithmic complexity can usher in pronounced differences in computational durations and resource utilization. This inherent intricacy prompts introspection: Is the potency of a given algorithm in addressing diverse practical quandaries commensurate with its inherent complexity. By juxtaposing the KL-UCB and UCB algorithms, this study not only highlights their relative merits and demerits but also furnishes insights that could serve as catalysts for further refinement and optimization. The overarching aim is to cultivate an informed perspective, guiding practitioners in choosing or fine-tuning algorithms tailored to specific applications without incurring undue computational overheads.

Movie Recommendation System Using KNN, Cosine Similarity and Collaborative Filtering

The movie recommendation system is becoming increasingly popular in the digital era. With the continuous emergence of a vast amount of movie resources, users are facing more and more choices. Therefore, an intelligent movie recommendation system can assist users in quickly finding movies that match their personal preferences, thereby enhancing user satisfaction and movie-watching experience . Our movie recommendation system recommends high-rated and well-reviewed films and TV shows to the general audience based on Netflix viewers’ ratings for them. These are the movies and shows that are considered worth watching. We utilize the Netflix open dataset, which consists of a large number of user ratings and movie information [1]. By merging the movie and rating datasets and performing data cleansing and preprocessing, we obtain a dataset suitable for models. Subsequently, we represent the movie rating data using sparse matrices and train a k-nearest neighbours (k-NN) algorithm to achieve personalized movie recommendations [2].Our approach considers both user rating preferences and movie similarities, enabling accurate and personalized recommendations for users. Through evaluation experiments and user surveys, we validate the effectiveness of our recommendation system in terms of accuracy and user satisfaction [3]. This research is significant for improving the effectiveness of movie recommendation systems and enhancing user experiences.

How consumers evaluate movies on online platforms? Investigating the role of consumer engagement and external engagement

This study examines how the causal patterns of consumers’ engagement and experts’ external engagement are associated with their online evaluation of movies. To this end, this study identifies the interplay of two dimensions (personal and interactive engagement) of consumers’ engagement and external engagement in their evaluation of movies and offers four propositions. This study utilizes fuzzy-set qualitative comparative analysis (fsQCA) on an original dataset of 264 movies released during 2018–2021 to test these propositions. Findings indicate seven configurations of consumers’ engagement behavior and external engagement that explain high movie ratings. This study presents fresh insights into how two dimensions of consumer engagement behavior and external engagement combine to explain movie ratings better. The findings accentuate the importance of synergies between dimensions of consumer engagement behavior and external engagement in driving the evaluation of movies.

Exploring the relationship between users characteristics and movie recommendations using a KNN-based recommender system

The central aim of this research is to elucidate the degree to which demographic variables, including but not limited to age and gender, bear on the performance of movie recommendation algorithms within film recommendation systems. Further, we endeavor to uncover any existent correlations between these user characteristics and the resultant outputs of such systems. Leveraging the expansive dataset available via MovieLens, we employ a linear regression model to ascertain the four critical variables (age, gender, occupation, and the average user rating for films previously watched) that have the most profound influence on movie recommendation algorithms. Once these salient factors have been determined, we assign their respective weights and incorporate these into a KNN algorithm. We then subject the resultant model to rigorous testing to verify the accuracy of our results and to ascertain whether the integration of these weighted elements enhances the overall precision of the movie recommendation system. While extant literature predominantly focuses on the amalgamation of KNN with other algorithms, our study charts a novel course by using linear regression. This methodology allows us to intuitively illustrate the relationship between user demographics and the movie recommendation system and enables us to evaluate whether emphasizing certain characteristics can augment the system's effectiveness. Our findings suggest that of all the user characteristics examined, the mean of users ratings for movies previously watched exerts the greatest influence on the outputs of the movie recommendation system. Moreover, incorporating weights reflective of the average user ratings across all movie features within the KNN algorithm can significantly bolster the accuracy of the resultant movie recommendations.

Geographical Tweets Based Current Affairs Prediction Using Hybrid Features

Twitter serves as a prominent and freely accessible social networking platform, enabling registered and authorized users to share their opinions and reviews through concise messages known as tweets. This research aims to conduct sentiment analysis on a Twitter dataset, seeking to predict and analyze current affairs based on user behaviour and opinions. The scope of current affairs encompasses a wide range of topics, including product reviews, political discourse, movie ratings, and more, expressed in both real-time streaming and offline data on the Twitter platform. During the feature extraction phase, a hybrid approach is employed, incorporating features such as acronyms, synonyms, and emoticons utilized by users in their tweets. Additionally, a hybrid dictionary is introduced, and instead of relying solely on unigrams and bigrams, a novel algorithm, n-gram, is implemented for more comprehensive analysis. This paper delves into various methodologies, specifically focusing on multiclass classification, and introduces a proposed system along with its noteworthy results. The research contributes to the understanding of user sentiments and behaviours on Twitter, offering valuable insights into diverse current affairs.

Movies Rating Prediction using Supervised Machine Learning Techniques

Social media are the most enormous and fast data technology on the internet. A massive amount of data is generated from the internet day by day. The efficient processing of such massive records is hard, so we require a system that learns from these facts and makes a useful prediction like machine learning. Machine learning strategies make the systems learn it. Applying K-nearest neighbors (KNN), Support Vector Machine (SVM), as well as Random Forest Model (RF), three supervised machine learning approaches, we attempted to develop a model for a movie recommendation system. This research not only compares the three models but also assesses how they compare favorably against one another in the areas of accuracy, accuracy, recalls, and F1 score. First of all, we discuss a summary of machine learning techniques used for recommendations of items in the past. We subsequently detail the most effective machine learning model for making predictions and recommendations about movie ratings. Predictions of movie reviews are made using K-nearest neighbor (KNN) and Support Vector Machine (SVM) models as well as the Random Forest Model (RF). The efficacy of all models was tested with Movie Lens 100k and 1M dataset. The random forest Model performs better as compared to other models. Furthermore, we summarize the current challenges based on state of art approaches. Finally, we present open issues and future directions for further research.

Predicting IMDb Rating of Movies by Machine Learning Techniques

Film Industry is not only a industry or a centre of entertainment, rather it is now a centre of global business. All over the world is now excited about a movie’s box office success, popularity etc. A huge data is available online about these movies success or popularity. We have used Hollywood movie list from Wikipedia and their rating from IMDb movie rating website to create our data set. Then machine learning classification algorithms are applied of the data set. Lastly an efficient model is developed to predict a movie’s IMDb rating. The model gives good classification measures with the data set. Index Terms-Movie Rating, Machine Learning, Prediction, Box Office, IMDb.

Research on the Impact of Parental Movie Choices on Children in China

In the context of growing global awareness regarding the influence of media content on young audiences, this paper focuses on the significant subject of film rating systems. Specifically, it explores the intricate landscape of these systems, their implementation across various countries, and the unique absence of such a system in China. Through an extensive analysis of existing rating systems, including those in the United States, the UK, and India, the study illuminates their influence on global cinema and cultural perceptions. Focusing on China, the paper employs a detailed survey of parents experiences and opinions regarding movie selection for children in the absence of a formal rating system. Findings reveal that a majority of parents acknowledge that movies have an impact on children. Notably, parents express a strong desire for the implementation of a movie rating system in China to guide their choices and prevent the exposure of their children to adult or violent content. However, the study also uncovers potential challenges and complexities in designing a system that balances protection, choice, and cultural relevance. The paper concludes by advocating for a well-considered film rating system in China that aligns with international norms yet is sensitive to local values, aiming to foster a responsible and diverse cinematic culture.

Movie recommendation system using taymon optimized deep learning network

The increased usage of the internet and social networks generates a large volume of information. Exploring through the large collection is time-consuming and hard to find the required one, so there is a serious need for a recommendation system. Based on this context several movie recommendation (MR) systems have been recently established. In addition, they have poor data analytics capability and cannot handle changing user preferences. As a result, there are many movies listed on the recommendation page, which provides for a poor user experience is the major issue. Therefore, in this work, a novel Taymon Optimized Deep Learning network (TODL net) for recommending top best movies based on their past choices, behaviour and movie contents. The deep neural network is a combination of Dilated CNN with Bi-directional LSTM. The DiCNN-BiLSTM model eliminates the functionality pooling operations and uses a dilated convolution layer to address the issue of information loss. The DiCNN is employed to learn the movie contents by mining user behavioral pattern attributes. The BiLSTM is applied to recommend the best movies on basis of the extracted features of the movie rating sequences of users in other social mediums. Moreover, for providing better results the DiCNN-BiLSTM is optimized with Taymon optimization algorithm to recommend best movies for the users. The proposed TODL net obtains the overall accuracy of 97.24% for best movies recommendation by using TMDB and MovieLens datasets.

Analyze IMDb movies by sentiment and topic analysis

Movie is an important cultural form, carrying multiple levels and meanings such as art, entertainment and social value. Movie review and rating data sets are huge, and deep learning and natural language processing methods are widely used today. Advances in big data and deep learning offer unprecedented opportunities to understand moviegoer behavior and preferences while providing a cost-effective way to gain insights relevant to the entertainment industry. This project conducts sentiment analysis, topic modeling, and visual statistical analysis based on the IMDb movie data set to identify key factors and deeper insights that influence successful decision-making in film production. This project first uses the word embedding method to vectorize the movie review text, and then uses Bidirectional Long Short-Term Memory (Bi-LSTM) to perform sentiment classification. In addition, statistical methods such as visualization were used to discover conclusions such as the highest average number of movies released in November, and identify trends, patterns and relationships between the variables of IMDb movies. Finally, the Latent Dirichlet Allocation (LDA) topic modeling model was constructed to find out that the important topic with increased demand is light entertainment movies, highlighting the commercial feasibility of comedy movies as a profitable business model. In summary, this project uses an emotion-topic fusion analysis method based on the Bi-LSTM emotion classification method and the LDA topic modeling method. The results show that the Bi-LSTM model can better identify positive and negative emotions in movie reviews, and the LDA topic model performs well in mining popular topics.

Personalized movie rating prediction based on the data of short video users and videos information

This study presents a sophisticated personalized movie rating prediction model that incorporates user-generated content from Douyin, a prominent video-based social media platform. The proposed model effectively leverages user demographic data, movie metadata, historical rating records, and sentiment analysis of user comments to precisely predict a user's rating score for films. This study employs lexicon-integrated two-channel CNN-LSTM family models for conducting sentiment analysis. Furthermore, the study explores a variety of regression models, such as linear regression, ridge regression, LASSO regression, and Elastic Net regression, to determine the most suitable model for the dataset being analyzed. This analysis aims to enhance the accuracy and effectiveness of the sentiment analysis performed on the given dataset. Furthermore, we contemplate the application of collaborative filtering algorithms, such as Alternating Least Squares (ALS), in developing our personalized movie rating prediction model. By incorporating user-generated content from Douyin, our methodology enhances the accuracy of movie rating predictions and underscores the significance of personalized recommendations in the era of social media.

A collaborative filtering recommendation framework utilizing social networks

Collaborative filtering is a widely used technique for providing personalized recommendations to users. However, traditional collaborative filtering methods fail to consider the social connections between users. The current study proposes a collaborative filtering recommendation framework that employs social networks to generate more precise and pertinent recommendations. The framework is based on a modified version of the user-based collaborative filtering algorithm, which computes user similarity based on their ratings and social connections. The similarity measure is determined by a weighted combination of these two factors, with the weights learned through an optimization process. The framework is evaluated using a dataset of movie ratings and social connections between users. The findings reveal that the proposed approach outperforms traditional collaborative filtering methods regarding recommendation accuracy and relevance. Moreover, the framework can offer more diverse recommendations compared to traditional methods. In summary, the proposed framework integrates social networks to enhance the accuracy and relevance of collaborative filtering recommendations. The approach has various applications, including e-commerce, music, and movie recommendation, and can potentially address the issues of cold-start and sparsity in collaborative filtering.

Movie Recommender System Using Decision Tree Method

In this modern era, many things that can be done online, one of which is watching movies. When the number of movies increases, people often find it difficult to decide which movie to watch next. To solve this problem, a useful recommendation system was developed to find movies that one might like based on movies that have been watched before. This research develops a movie recommendation system using Collaborative Filtering (CF) with the Decision Tree algorithm. In this study, the data used were movie data and ratings obtained from the grouplens.org website. Then the movielens dataset is filtered and only saves movies with a rating of more than 50 that are used in the recommendation system. In this study, Mean Absolute Error (MAE) is used as a method to assess the accuracy of the movie recommendation system. Based on the research that has been done, Decision Tree gets better results with an MAE value of 0,942 compared to Collaborative Filtering with an MAE value of 1,242.