Movie Domain Research Articles

Prediction techniques of movie box office using neural networks and emotional mining

Box office prediction is of great significance for understanding investment risks, class construction, promotion and distribution, and theater scheduling. However, due to the insufficient selection of influencing factors of movie box office, the currently existing prediction model restricts the prediction accuracy. A total of 34 influencing factors in 11 categories, such as heat index, movie types, release date, creators, first-day box office, were selected to study the prediction technology of movie box office. The Word2vec algorithm is used to construct a feature thesaurus for nouns in movie domain; adjectives and verbs with emotional coloring are used to construct an emotional dictionary based on the movie domain; and the TF-IDF algorithm is integrated to calculate the emotional scores of movie comments. A prediction method based on comments and Multivariate Linear Regression (MLR) is designed to analyze the relationship between the influencing factors and the movie box office, which provides an important basis for the prediction of the total box office, and also provides a decision-making reference for the movie industry and the related management departments. Incorporating comments as feature values to improve the accuracy, a prediction model based on comments and Convolutional Neural Network (CNN) is constructed. The results show that the average prediction accuracy of the MLR without comments, Back-Propagation Neural Network (BPNN), and CNN is 63.4%, 68.3%, and 71.9%, respectively, and after integrating the comments, the average prediction accuracy of the MLR and CNN is improved by 16.1% and 11.8%, respectively, and the prediction accuracy is significantly improved.

X-Model4Rec: An Extensible Recommender Model Based on the User’s Dynamic Taste Profile

Several approaches have been proposed to obtain successful models to solve complex next-item recommendation problem in non-prohibitive computational time, such as by using heuristics, designing architectures, and applying information filtering techniques. In the current technological scenario of artificial intelligence, sequential recommender systems have been gaining attention and they are a highly demanding research area, especially using deep learning in their development. Our research focuses on an efficient and practical model for managing sequential session-based recommendations of specific products for users using the wine and movie domains as case studies. Through an innovative recommender model called X-Model4Rec – eXtensible Model for Recommendation, we explore the user's dynamic taste profile using architectures with transformer and multi-head attention mechanisms to solve the next-item recommendation problem. The performance of the proposed model is compared to that of classical and baseline recommender models on two real-world datasets of wines and movies, and the results are better for most of the evaluation metrics.

Enhanced Content-Based Recommendation Using Topic Modelling and Knowledge Graph

Content-based (CB) recommendation algorithms recommend items to a user based on items the user liked in the past. CB methodologies have gained attention due to their higher accuracy and transparency and the emergence of new technologies, such as knowledge graphs (KGs), advances in natural language processing (NLP), and sentiment analysis. While most previous studies have mainly focussed on the use of term frequency-inverse document frequency (TF-IDF) and other related enhancements, little work can be found on using KGs in CB recommendations. This paper presents an enhancement of the conventional CB recommendation by incorporating KGs for a movie domain. The graph is constructed using the MovieLens data set, which is augmented with additional features such as actors, directors, and genres. Furthermore, the graph is expanded by incorporating topics derived from latent dirichlet allocation (LDA) extraction. Using the KGs, the proposed approach enhances user profiles by leveraging the interconnected user-movie relationship within a graph structure. The results of the experiments showed that the proposed approach exceeded the tested baselines in terms of precision, recall, and F-score metrics.

An approach for explaining group recommendations based on negotiation information

Explaining group recommendations has gained importance over the last years. Although the topic of recommendation explanation has received attention in the context of single-user recommendations, only a few group recommender systems (GRS) currently provide explanations for their group recommendations. However, those GRS that support explanations, provide either explanations being highly reliant on the aggregation technique used for generating the recommendation (most of them trying to tackle shortcomings of the underlying technique), or explanations with a rich content but requiring users to provide considerable additional data. In this article, we present a novel approach for providing explanations of group recommendations, which are generated by a GRS based on multi-agent negotiation techniques. An evaluation of our approach with a user study in the movies domain has shown promising results. Explanations provided by our GRS system helped users during the decision-making process, since they modified the feedback given to recommended items. This is an improvement with respect to systems that do not provide explanations for their recommendations.

Tell me what you Like: introducing natural language preference elicitation strategies in a virtual assistant for the movie domain

Tell me what you Like: introducing natural language preference elicitation strategies in a virtual assistant for the movie domain

Automatic assignment of microgenres to movies using a word embedding-based approach

Streaming services are increasingly leveraging Artificial Intelligence (AI) technologies for improved content cataloging, user experiences in content discovery, and personalization. A significant challenge in this domain is the automated assignment of microgenres to movies. This study introduces and evaluates approaches based on clustering, topic modeling, and word embedding to address this task. The evaluation employs a preprocessed dataset containing movie-related data—title tags, synopses, genres, and reviews—alongside a predefined microgenre list. Comparisons of three activation functions (binary step, ramp, and sigmoid) gauge their effectiveness in augmenting microgenre tags. Results demonstrate the superiority of the word embedding approach over clustering and topic modeling in terms of mean accuracy. Even more, the word embedding approach stands as the sole fully automated solution. Analysis indicates that incorporating review-based tags introduces noise and undermines accuracy. Besides, the word embedding approach yields optimal outcomes using the sigmoid function, effectively doubling assigned tags while maintaining matching quality. This sheds light on the potential of word embedding methods within the movie domain.

A Bi-Directional GRU Architecture for the Self-Attention Mechanism: An Adaptable, Multi-Layered Approach with Blend of Word Embedding

Sentiment analysis (SA) has become an essential component of natural language processing (NLP) with numerous practical applications to understanding “what other people think”. Various techniques have been developed to tackle SA using deep learning (DL); however, current research lacks comprehensive strategies incorporating multiple-word embeddings. This study proposes a self-attention mechanism that leverages DL and involves the contextual integration of word embedding with a time-dispersed bidirectional gated recurrent unit (Bi-GRU). This work employs word embedding approaches GloVe, word2vec, and fastText to achieve better predictive capabilities. By integrating these techniques, the study aims to improve the classifier’s capability to precisely analyze and categorize sentiments in textual data from the domain of movies. The investigation seeks to enhance the classifier’s performance in NLP tasks by addressing the challenges of underfitting and overfitting in DL. To evaluate the model’s effectiveness, an openly available IMDb dataset was utilized, achieving a remarkable testing accuracy of 99.70%.

Automatic assignment of moral foundations to movies by word embedding

Morality is a topic that people are increasingly concerned about. Morality is observed and measured during public acts or when developing and consuming products, such as movies. The Moral Foundations Theory (MFT) was developed to rigorously perform these measurements with the support of the Moral Foundations Dictionary (MFD). In this paper, a Word Embedding-based Moral Foundation Assignment (WEMFA) approach has been designed, implemented, and applied to the movie domain for multiple assignment of moral foundations. WEMFA may use any dictionary, and it has been applied to a movie collection generated from movie synopses. A comparison between WEMFA and MoralStrength, the only approach found in the scientific literature, has been carried out. The proposed approach provided a percentage improvement of 41.7% with respect to the best version of MoralStrength, which uses an extension of the original MFD almost 10 times larger in number of terms. In addition, an extension of the original MFD (MFD24) has been built by adding 14 new moral foundations to the 10 original ones, enriching the moral context. WEMFA provided a mean accuracy of 78% with MFD24 despite the increment of the number of moral foundations. Besides, new extended dictionaries or even totally different ones can be used with WEMFA, since it does not need any training.

Parallel Split-Join Networks for Shared Account Cross-Domain Sequential Recommendations

Sequential recommendation is a task in which one models and uses sequential information about user behavior for recommendation purposes. We study sequential recommendation in a context in which multiple individual users share a single account (i.e., they have a shared account) and in which user behavior is available in multiple domains (i.e., recommendations are cross-domain). These two characteristics bring new challenges on top of those of the traditional sequential recommendation task. First, we need to identify the behavior associated with different users and different user roles under the same account in order to recommend the right item to the right user role at the right time. Second, we need to identify behavior in one domain that might be helpful to improve recommendations in other domains. We study <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">shared account cross-domain sequential recommendation</i> and propose a <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p</b> arallel <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">s</b> plit- <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">j</b> oin <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Net</b> work (Parallel Split-Join Network (PSJNet)), a parallel modeling network to address the two challenges above. We use “split” to address the challenge raised by shared accounts; PSJNet learns role-specific representations and uses a gating mechanism to filter out, from mixed user behavior, information of user roles that might be useful for another domain. In addition, “join” is used to address the challenge raised by the cross-domain setting; PSJNet learns cross-domain representations by combining the information from “split” and then transforms it to another domain. We present two variants of PSJNet: PSJNet-I and PSJNet-II. PSJNet-I is a “split-by-join” framework that splits the mixed representations to get role-specific representations and joins them to obtain cross-domain representations at each timestamp simultaneously. PSJNet-II is a “split-and-join” framework that first splits role-specific representations at each timestamp, and then the representations from all timestamps and all roles are joined to obtain cross-domain representations. We concatenate the in-domain and cross-domain representations to compute a recommendation score for each item. Both PSJNet-I and PSJNet-II can simultaneously generate recommendations for two domains where user behavior in two domains is synchronously shared at each timestamp. We use two datasets to assess the effectiveness of PSJNet. The first dataset is a simulated shared account cross-domain sequential recommendation dataset obtained by randomly merging the Amazon logs from different users in the movie and book domains. The second dataset is a real-world shared account cross-domain sequential recommendation dataset built from smart TV watching logs of a commercial organization. Our experimental results demonstrate that PSJNet outperforms state-of-the-art sequential recommendation baselines in terms of MRR and Recall.

Rating-Based Recommender System Based on Textual Reviews Using IoT Smart Devices

Recommender system (RS) is a unique type of information clarification system that anticipates the user's evaluation of items from a large pool based on the expectations of a single stakeholder. The proposed system is highly useful for getting expected meaning suggestions and guidance for choosing the proper product using artificial intelligence and IoT (Internet of Things) such as chatbot. The current proposed technique makes it easier for stakeholders to make context-based decisions that are optimal rather than reactive, such as which product to buy, news classification based on high filtering views, highly recommended wanted music to choose, and desired product to choose. Recommendation systems are a critical tool for obtaining verified information and making accurate decisions. As a result, operational efficiency would skyrocket, and the risk to the company that uses a recommender system would plummet. This proposed solution can be used in a variety of applications such as commercial hotels OYO and other hotels, hospitals (GYAN), public administrative applications banks HDFC, and ICICI to address potential questions on the spot using intelligence computing as a recommendation system. The existing RS is considering a few factors such as buying records, classification or clustering items, and user's geographic location. Collaborative filtering algorithms (CFAs) are much more common approaches for cooperating to mesh the respective documents they retrieved from the historical data. CFAs are distinguished in plenty of features that are uncommon from other algorithms. In this existing system classification, precision and efficiency and error rate are statistical measurements that need to be enhanced according to the current need to fit for global requirements. The proposed work deals with enhancing accuracy levels of text reviews with the recommender system while interacting by the numerous users for their domains. The authors implemented the recommender system using a user-based CF method and presented the significance of collaborative filtering on the movie domain with a recommender system. This whole experiment has been implanted using the RapidMiner Java-based tool. Results have been compared with existing algorithms to differentiate the efficiency of the current proposed approach.

Balancing the trade-off between accuracy and diversity in recommender systems with personalized explanations based on Linked Open Data

Collaborative filtering recommendation algorithms generate suggestions based on similar interactions between users. Although it provides accurate recommendations, the approach has two limitations: the popularity bias, which frequently suggests a small set of the most interacted items, and the systems’ black box functioning, as they are grounded on complex mathematical models. To improve such aspects in collaborative filtering algorithms, this paper introduces a multi-domain item reordering system based on the best explanation for an item, which are the best ranked paths extracted from a Linked Open Data knowledge graph connecting recommended and interacted items. To order paths, the algorithm assigns a value to the node attributes connecting two items by calculating the popularity of the property between interacted items that are rare in the full set of items. Results from two datasets of the movie and music domains comparing the proposed reordering system with six baselines of different collaborative filtering families showed that our easy-to-explain approach improved diversity and/or accuracy metrics.

Offline Evaluation of Recommender Systems in a User Interface With Multiple Carousels.

Many video-on-demand and music streaming services provide the user with a page consisting of several recommendation lists, i.e., widgets or swipeable carousels, each built with specific criteria (e.g., most recent, TV series, etc.). Finding efficient strategies to select which carousels to display is an active research topic of great industrial interest. In this setting, the overall quality of the recommendations of a new algorithm cannot be assessed by measuring solely its individual recommendation quality. Rather, it should be evaluated in a context where other recommendation lists are already available, to account for how they complement each other. The traditional offline evaluation protocol however does not take this into account. To address this limitation, we propose an offline evaluation protocol for a carousel setting in which the recommendation quality of a model is measured by how much it improves upon that of an already available set of carousels. We also propose to extend ranking metrics to the two-dimensional carousel setting in order to account for a known position bias, i.e., users will not explore the lists sequentially, but rather concentrate on the top-left corner of the screen. Finally, we describe and evaluate two strategies for the ranking of carousels in a scenario where the technique used to generate the two-dimensional layout is agnostic on the algorithms used to generate each carousel. We report experiments on publicly available datasets in the movie domain to show how the relative effectiveness of several recommendation models compares. Our results indicate that under a carousel setting the ranking of the algorithms changes sometimes significantly. Furthermore, when selecting the optimal carousel layout accounting for the two dimensional layout of the user interface leads to very different selections.

Active tag recommendation for interactive entity search: Interaction effectiveness and retrieval performance

We introduce active tag recommendation for interactive entity search, an approach that actively learns to suggest tags from preceding user interactions with the recommended tags. The approach utilizes an online reinforcement learning model and observes user interactions on the recommended tags to reward or penalize the model. Active tag recommendation is implemented as part of a realistic search engine indexing a large collection of movie data. The approach is evaluated in task-based user experiments comparing a complete search system enhanced with active tag recommendation to a control system in which active tag recommendation is not available. In the experiment, participants (N = 45) performed search tasks on the movie domain and the corresponding search interactions, information selections, and entity rankings were logged and analyzed. The results show that active tag recommendation (1) improves the ranking of entities compared to written-query interaction, (2) increases the amount of interaction and effectiveness of interactions to rank entities that end up being selected in a task, and (3) reduces, but does not substitute, the need for written-query interaction (4) without compromising task execution time. The results imply that active learning for search support can help users to interact with entity search systems by reducing the need for writing queries and improve search outcomes without compromising the time used for searching.

Benchmarking Fairness Measures for Calibrated Recommendation Systems on Movies Domain

Benchmarking Fairness Measures for Calibrated Recommendation Systems on Movies Domain

Addressing the Cold-Start Problem in Collaborative Filtering Through Positive-Unlabeled Learning and Multi-Target Prediction

The cold-start problem is one of the main challenges in recommender systems and specifically in collaborative filtering methods. Such methods, albeit effective, typically can not handle new items or users that do not have any prior interaction activity in the system. In this paper, we propose a novel two-step approach to address the cold-start problem. First, we view the user-item interactions in a positive unlabeled (PU) learning setting and reconstruct the interaction matrix between users and warm items, detecting missing links and recommending warm items to existing users. Second, an inductive multi-target regressor is trained on this reconstructed interaction matrix and subsequently predicts interactions for new items that enter the system. To the best of our knowledge, this is the first time that such a two-step PU learning method is proposed to address the cold-start problem in recommender systems. To evaluate the proposed approach, we employed four benchmark datasets from movie and news recommendation domains with explicit and implicit feedback. We compared our method against three other competitor approaches that address the cold-start problem and showed that our proposed method significantly outperforms them, achieving in a case an increase of 16.9% in terms of NDCG.

Explaining recommender systems fairness and accuracy through the lens of data characteristics

The impact of data characteristics on the performance of classical recommender systems has been recently investigated and produced fruitful results about the relationship they have with recommendation accuracy. This work provides a systematic study on the impact of broadly chosen data characteristics (DCs) of recommender systems. This is applied to the accuracy and fairness of several variations of CF recommendation models. We focus on a suite of DCs that capture properties about the structure of the user–item interaction matrix, the rating frequency, item properties, or the distribution of rating values. Experimental validation of the proposed system involved large-scale experiments by performing 23,400 recommendation simulations on three real-world datasets in the movie (ML-100K and ML-1M) and book domains (BookCrossing). The validation results show that the investigated DCs in some cases can have up to 90% of explanatory power – on several variations of classical CF algorithms –, while they can explain – in the best case – about 40% of fairness results (measured according to user gender and age sensitive attributes). Therefore, this work evidences that it is more difficult to explain variations in performance when dealing with fairness dimension than accuracy.

Representing emotions with knowledge graphs for movie recommendations

Consumption of media, and movies in particular, is increasing and is influenced by a number of factors. One important and overlooked factor that affects the media consumption choices is the emotional state of the user and the decision making based on it. To include this factor in movie recommendation processes, we propose a knowledge graph representing human emotions in the domain of movies. The knowledge graph has been built by extracting emotions out of pre-existing movie reviews using machine learning techniques. To show how the knowledge graph can be used, a chatbot prototype has been developed. The chatbot’s reasoning mechanism derives movie recommendations for the user by combining the user’s emotions, which have been extracted from chat messages, with the knowledge graph. The developed approach for movie recommendations based on sentiment represented as a knowledge graph has been proven to be technically feasible, however, it requires more information about the emotions associated with the movies than currently available online.

Exploiting personalized calibration and metrics for fairness recommendation

Recommendation systems are used to suggest items that users can be interested in. These systems are based on the user preference historic to create a recommendation list with items that have the higher similarity with the user’s interests, in order to achieve the best possible user’s satisfaction, which is usually measured as recommendation precision. However, the search for the best precision can cause some side effects such as overspecialization, few diversity and miscalibration of genres, classes and niches. Calibration provides fairer recommendations, which respect the genre proportionality on the user’s preferences, avoiding overspecialization. This article aims to explore ways to balance the trade-off weight between precision and calibration based on divergence measures, as well as to propose metrics to evaluate the calibration in the suggested list. The proposed system works in a post-processing step and does not depend on a specific recommender algorithm or workflow. For this purpose, we evaluate six recommender algorithms applied in the movie domain, analyzing variations of three fairness measures, two personalized trade-off weights and eleven constant weights. To understand the results we use the precision, the reciprocal rank and two proposed metrics. The results indicate that the trade-off formulation of personalized weights obtains better results when used to compare the recommendation lists using matrix factorization-based approaches on Movielens dataset. In addition, the calibration also impacts the precision and fairness of all considered algorithms used in evaluation.

Extending collaborative filtering recommendation using word embedding: A hybrid approach

SummaryCollaborative filtering recommendation systems, which analyze sets of user ratings, have been applied to various domains and have resulted in considerable improvements in the traditional recommendation system. However, they still have problems with data sparsity and cold‐start of the user ratings. To solve these problems, we present a hybrid recommendation approach by combining collaborative filtering methods and word embedding‐based content analysis. This study focuses on the movie domain, and therefore, the contents of the items are represented as a set of features such as titles, genres, directors, actors, and plots. The main aim of this paper is to understand the content of the movie plot using a word embedding to improve the measurement of similarity of each plot content to other plot content (called plot embedding). To enhance the accuracy in measuring the similarity between movies, we also consider other features such as titles, genres, directors, and actors extracted from movies. In the experiments, the movie dataset was collected by our crowdsourcing platform, which is the OMS platform. The experimental findings indicate that the proposed approach can enhance the efficiency of applied collaborative filtering recommendation systems.

An efficient sentiment analysis using topic model based optimized recurrent neural network

Abstract In recent years, topic modeling and deep neural network-based methods have attracted much attention in sentiment analysis of online reviews. This paper presents a hybrid topic model-based approach for aspect extraction and sentiment classification of textual reviews. Latent Dirichlet allocation applied for aspect extraction and two-layer bi-directional long short-term memory (LSTM) for sentiment classification. This work also proposes a hill climbing-based approach for tunning model hyperparameters. The proposed model evaluated on three different datasets. Compared to the single-layer Bi-LSTM model, the proposed model gives 95, 95, and 86% accuracy for the movie, mobile, and hotel domain, respectively.