302260center
s302260center8/21/2023
302260center
28654381370648(NEF3001) Applied Project 1 -Project Proposal
Sentiment Based Movie Rating System
2865438137064831448389405938Adarsha Bhandari(s4675648), Yifan Sheng(s4584372), Pranav Dhakal(s4643813)
VU SYDNEY
31448389405938
Sentiment Based Movie Rating System
Table of contents
Project Background and Description 2
Project Scope 5
Functional Requirements 6
Non-functional Requirements 8
Use case 10
Sequence Diagram .. 15
Flow Chart 18
User Interface Design 19
Resource Management Plan 20
Risk Management Plan 21
Milestones 22
Conclusion 23
Contribution 24
Reference 25
Figures and tables
Figure 5.1 Use Case Diagram 10
Figure 6.1 Register New User 15
Figure 6.2 Login 16
Figure 6.3 Search for Movies 16
Figure 6.4 Adding Comments on the Movie 17
Figure 6.5 Editing Comments on the Movie 17
Figure 7.1 Flow Chart 18
Figure 8.1 User Interface Design 19
Figure 11.1 Development schedule 22
Figure 11.2 Development schedule in Gantt Chart 23
Table 5.1 Use Case Register 11
Table 5.2 Use Case Login 12
Table 5.3 Use Case Search Movies 13
Table 5.4 Use Case Enter Comment 14
Table 10.1 Risk management table 21Table 13.1 Contribution table 24
Project Background and Description s4643813 Pranav Dhakal
Introduction:
The Sentiment Based Movie Rating System is a truly avant-garde tool meticulously designed to decipher and unravel the complex and multifaceted tapestry of audience sentiment in relation to cinema. Despite the influx of films flooding the cinematic landscape and the kaleidoscope of opinions accompanying them, deciphering the authentic pulse of the audience proves an intricate task. This groundbreaking system stands as an exemplar solution poised to tackle this conundrum, harnessing state-of-the-art technology not only to ascertain the binary preference of liking, or disliking a movie but also to extrapolate the intricate spectrum of emotions that intertwine with these cinematic experiences.
With the Sentiment Based Movie Rating System, conventional paradigms of audience appraisal are transcended, at the cutting edge of technological innovation. This innovative system shifts paradigms away from conventional ratings systems that rely on quantifiable metrics like stars or numerical values. It explores the depths of textual reviews crafted by individuals, dissecting these narratives to discern the rich undercurrents of emotions that permeate their discussion of movies. The system employs advanced analytical methodologies to unravel the complex sentiments and nuanced viewpoints encapsulated within these reviews.
Essentially, the Sentiment Based Movie Rating System represents a pioneering quest to unravel the enigmatic context of audiences' sentiments. As a result, it offers insight beyond the usual scoring mechanism into the emotional complexity of cinematic encounters. Through meticulous analysis of textual criticism and the multitude of emotions interwoven within them, the system wishes to shed light on the variety of emotions that films evoke in audiences. Performing sentiment analysis in the context of a Sentiment-Based Movie Rating System involves a multifaceted approach that leverages natural language processing (NLP) techniques, machine learning algorithms, and analytics Meaning. Here is a high-level overview of how sentiment analysis can be performed:
1) Data collection and preprocessing:
The system starts by collecting text-based movie reviews from a variety of sources such as websites, social media platforms, and user submissions. These revisions are then preprocessed to remove noise, including special characters, punctuation, and stop words. Text is denoted by words or sentences, allowing for further analysis.
2) Glossary and dictionary of emotions:
An emotional vocabulary or dictionary containing words related to a particular emotion is used. Each word is assigned a sentiment score, usually from negative to positive. These vocabulary words help determine the overall opinion of the review by aggregating individual word scores.
3) Machine learning models:
Supervised machine learning models can be used to classify reviews into positive, negative, or neutral categories. These models are trained on labeled data where reviews are manually annotated with sentiment labels. Popular algorithms like Naive Bayes, Support Vector Machines, and deep learning models like Recurrent Neural Networks (RNN) or Transformers can be used for this task.
4) Emotion detection:
To capture emotional nuances, more advanced methods are deployed. Emotion detection models can identify a wide range of emotions in text, such as happiness, anger, sadness, and more. These models are typically trained on labeled datasets that contain text annotated with specific emotional labels.
5) Context analysis:
Sentiment analysis also takes into account the context in which words are used. Negations and complements can dramatically change the mood of a sentence. Advanced models such as Transformers, especially BERT (Transformers Bidirectional Encoder Representation), have demonstrated proficient context understanding, thus improving the accuracy of sentiment analysis.
6) Hybrid method:
Combining methods that combine sentiment vocabularies, machine learning, and deep learning techniques can yield a more comprehensive understanding of sentiment. These methods can improve accuracy and accommodate the complexity of human language.
7) Real-time updates:To update movie ratings in real time based on new reviews, sentiment analysis is applied to newly submitted reviews. The calculated sentiment is then included in the overall rating of the film, flexibly reflecting the emotional evolution of the audience.
Ensuring privacy and security in a sentiment-based movie rating system is paramount to building user trust and protecting sensitive information. Here is an overview of strategies and measures that can be taken to address these concerns:
1) Data encrypt:
Use of encryption techniques such as SSL/TLS for secure communication between users and system servers. In addition, sensitive data, such as user credentials and personal information, must be encrypted when stored in the database.
2) User authentication and authorization:
Implementation of robust user authentication mechanisms, including multi-factor authentication (MFA), to verify the identity of registered users. Distinguishing user roles (e.g. registered vs registered). unregistered) and apply role-based access controls to ensure that only the right features can be accessed by each user.
3) Anonymization and aggregation:
Anonymize personally identifiable information (PII) in reviews and user profiles. Instead of storing explicit user details, combine reviews and ratings with anonymous identifiers. Aggregate data for analysis to help protect individual privacy while extracting valuable insights.
4) Privacy Policy and Consent:
clear communication of the system's privacy policies to users, describing how their data will be collected, used and protected. Obtain explicit user consent for data collection and processing activities, and provide users with the ability to withdraw consent and delete their data.
5) Regular security testing and penetration testing:
Perform periodic security audits and penetration testing to identify system vulnerabilities. Promptly remediate identified weaknesses to prevent potential breaches.
6) Safe development practices:
Following safe coding practices to reduce the risk of creating vulnerabilities during development. Regularly update software components and libraries to include the latest security patches.
7) Minimize data:
Collection of only the data necessary for the operation of the system. Minimize the collection of sensitive information and ensure that the data collected is used only for the intended purpose.
8) Secure storage and infrastructure:
Choosing a reputable hosting provider that offers strong security features. Use firewalls, intrusion detection systems, and other security measures to protect system infrastructure.
9) Moderate user-generated content:
Implementation of content moderation mechanisms to identify and filter inappropriate or malicious content from user-generated reviews and comments.
10) Regulatory compliance:
Ensuring compliance with relevant privacy and data protection regulations, such as GDPR, CCPA and HIPAA, depending on the jurisdiction and the data being processed.
11) Regular user training:
Educating users on security best practices, such as creating strong passwords, avoiding sharing personal information, and being aware of phishing attempts.
12) Data breach response plan:
This includes notifying affected users, stopping violations, and cooperating with authorities if necessary.By combining these measures, the Sentiment-Based Movie Rating System can create a safe and privacy-conscious environment, assuring users that their data is handled safely. responsibly and with extreme caution.
The discernment of audience sentiments within the cinematic realm holds multifarious implications. Filmmakers can gain profound insights into the reception of their creations, not merely limited to the polarity of approval or disapproval but extending into the very essence of emotions that are provoked. Studios and distributors can better tailor their marketing strategies and identify their target audience's emotional triggers. Moreover, audiences themselves can benefit from a more nuanced understanding of how their peers perceive movies, facilitating informed choices in their cinematic explorations. In short, the Sentiment Based Movie Rating System is an ingenious marvel at the intersection of technology and cinema. This system has pioneered a transformative approach to understanding audiences' emotions, encompassing the entire range of emotions that underlie their cinematic experience. Its use promises to reshape the way we view, critique, and interact with films, fostering deeper connections between filmmakers, audiences, and the complex tapestry of emotions that evocative movies.
2. Project Scope s4643813 Pranav Dhakal
To begin with, the system undertakes the task of gathering reviews and ratings from diverse sources, encompassing movie websites and social media platforms. Employing sophisticated sentiment analysis techniques, it discerns the inherent tone of these reviews, identifying whether they convey a positive, negative, or neutral viewpoint towards the movie. If one seeks insight into a specific film, accessing the system unveils a comprehensive overview, elucidating the degree of favour, disfavour, or neutrality expressed by the audience.However, the system's utility stretches beyond a surface-level presentation of sentiments. It provides
a comprehensive breakdown of emotions, illuminating the extent of audience elation, mild agitation, or melancholy experienced in response to the film. Ensuring heightened accuracy, the system continuously learns and evolves over time, adapting to comprehend the evolving intricacies of human language and expression. Essentially, this system revolves around the comprehension of individuals' cinematic viewpoints, derived directly from their own verbal articulations and ratings.
Importantly, the system refrains from attempting to predict overarching industry trends or engaging in intricate analyses. Instead, its primary focus remains steadfastly rooted in the audience's realm and their perceptions. Furthermore, the system's accessibility through standard web browsers ensures user-friendliness and seamless utilization. In the future, stakeholders such as filmmakers, studios, and those propelled by curiosity stand to derive substantial advantages from this system, as it affords an exclusive insight into the collective sentiments of the audience. As an anticipated future enhancement, the system could potentially offer real-time insights into audience sentiments upon a movie's release, providing a prompt and insightful gauge of initial reactions. In essence, the Sentiment Based Movie Rating System emerges as a departure from numerical evaluation, emphasizing the paramount significance of emotions within the cinematic landscape and offering an experience akin to engaging in a cinematic dialogue with acquaintances, augmented by advanced intelligence.
The following points are some of the project scopes.1) Data collection and sentiment analysis:
The main goal of the project was to design a sophisticated system capable of collecting movie reviews and ratings from a variety of sources, including movie websites and social media platforms. The system will use advanced sentiment analysis techniques to accurately identify the underlying feelings of these assessments, classifying them as positive, negative, or neutral.
2) Full audience overview:
Beyond simply determining sentiment, the project aims to provide users with a comprehensive understanding of audience reactions to movies. The system will present detailed analysis of emotions, highlighting different levels of emotional reactions such as excitement, restlessness or melancholy expressed by the audience when watching a particular movie.
3) Continuous learning and evolution:
To ensure the sustainability and accuracy of the system's sentiment analysis, the project includes implementing mechanisms for continuous learning. The system will evolve over time, adapting to changes in human language patterns and expressions, improving its ability to interpret and correctly classify sentiment in assessments.
4) User-friendly accessibility and future enhancements:
The project emphasizes user-friendliness and accessibility by providing the system through standard web browsers. In the future, potential improvements to the system include providing real-time feedback on audience emotions as soon as a movie is released. This feature can provide valuable and timely information to filmmakers and studios about initial audience reaction.
5) Audience-centric approach and industry applications:
The project emphasizes an audience-centric approach by limiting analysis of industry trends or engaging in complex analysis. Instead, it aims to provide a platform for individuals to share their perceptions and feelings regarding movies. Stakeholders, including filmmakers and studios, can leverage this system to better understand audience psychology, help refine marketing strategies, and improve the cinematic experience.
Some of the names of the related works are IMDB and Rotten Tomatoes, which are the top websites for sentiment based movie rating systems. People check these sites first just to know what to expect from a movie. 1) IMDb (Internet Movie Database):
IMDb is an extensive online database dedicated to movies, TV shows and celebrities. It serves as a comprehensive source of information, containing detailed information about movies, TV series, cast and crew members, production companies, release dates, plot summaries, trivia, etc Users can search for specific titles, explore genres, and access user-generated reviews and ratings. IMDb's User Ratings provide an aggregate measure of audience opinion, allowing users to gauge the overall reception of a movie or TV show.
2) Rotten tomatoes:
Rotten Tomatoes is a popular TV and film review aggregator, aggregating both critical reviews and ratings. It assigns a "Tomatometer" score to films based on a percentage of positive reviews from professional critics. The site classifies reviews as "Fresh" or "Rotten", indicating favorable or unfavorable reviews, respectively. In addition to the Tomatometer score, Rotten Tomatoes also gives an Audience score, which reflects how viewers rate the film. This dual perspective provides users with a comprehensive overview of audience and critical reception.
IMDb acts as an extensive online movie and TV database, providing comprehensive information and user-generated reviews. Rotten Tomatoes aggregates reviews from professional critics and audiences to calculate Tomatometer and audience scores, providing a dual perspective on film reception.
Some Articles related to our works :
Adarsha Bhandari (s4675648)
Sentiment analysis algorithms and applications A survey(Ravik, 2015)
- This study provides a thorough, up-to-date evaluation of the research conducted between 2002 and 2014 on numerous SA-related topics. Subjectivity classification, sentiment classification, review usefulness measurement, lexicon generation, opinion word and product aspect extraction, opinion spam detection, and diverse applications of opinion mining are the seven broad categories in which the article is discussed.
Improving aspect extraction by augmenting a frequency-based method with web-based similarity measures(Lis, 2015)
- In this work, a technique to enhance the effectiveness of aspect extraction from online customer reviews was developed. The technique uses PMI-IR to enhance frequency-based extraction. It also expands RCut to learn thresholds for deciding which candidate aspects to choose. Results from tests indicate that our suggested approach not only performs better than the most recent method for aspect extraction but also be generalizedgeneralised to diverse data sizes and product domains.
Feature based summarization of customers reviews of online products.(Bafnak, 2013)
- In this study, they suggested a cutting-edge method for dynamic feature-based summarizing of customer evaluations that operates in accordance with the product's domain. It was based on opinion mining and natural language processing. Results show that the suggested approaches accomplish their tasks very effectively and efficiently. Users may now quickly and effectively skim through the product reviews, making it much easier for them to absorb the information present in enormous amounts of product review content.
Deep learning for aspect-based sentiment analysis: A comparative review.(DoH.H, 2019)
- There have been active attempts to analyze, classify, comprehend, and anticipate the nature and opinion polarity of written languages to fine-grained levels with the emergence of user-generated content as a rich source of subjective information to do sentiment analysis at the aspect level. This work has offered a thorough review of the main deep learning algorithms and a detailed comparison of these approaches.
-
3. Functional Requirements
Adarsha bhandari(s4675648)
User verification and registration.
- To use the website's features, users must create an account and verify it by registering the account and verifying it.
- Users can log in using their login credentials after the accounts have been verified.
- Users can reset their passwords via email confirmation if they forget them.
- Users can delete the account through a similar process.
Browse information about movies.
- To find movies, users can perform a title, genre, or keyword search.
- After searching the name of the movie, movies that have been linked to the search appear on the loading screen.
- The system should provide movie recommendations based on user interests and viewing habits.
- Users can comment on the movies and edit the comments on the movies.
The Movie Database.
- The system must have a large movie database.
- Movies should be grouped together using genres.
- Detailed information on movies should be included, including the plot summary, cast, director, and release date.
User opinions and scores
- Movies can receive a star rating from 1 to 5 from registered users.
- Users can write and share their movie reviews.
-Users can adjust or delete their own ratings and reviews.
Sentiment Analysis
- The system must analyze user reviews for sentiment. The sentiment score (positive, negative, or neutral) for each review should be determined.
Aggregated Reviews
- The system must determine the overall ratings for each film based on user ratings.
-Aggregated reviews must be visible on movie pages.
User Profile
- Users can personalize their user accounts by adding avatars and personal information.
-Each user's rating history and preferred genres are visible in their profiles.
Admin Panel
- Administrators have access to user accounts.
- They can add to, amend, or remove the film database.
- Administrators have control over user ratings and reviews.
Notification system.
-Users receive notifications when new movies are launched, which movies are suggested, and when reviews are available.
Users' Interactions.
- Users can follow one another and communicate.
- Customers can leave comments and like reviews.
- Users can publish information, such as movie reviews.
Mobile -friendliness.
- The system should be adapted to the screens of mobile phones. (Shvetsova, 2023)
4. Non-Functional Requirements:
Adarsha bhandari(s4675648)
Performance:
- It is important as it will determine how quickly the system responds to user input (load times, for instance, should be under two seconds).
- Multiple users should be supported without performance being significantly affected.
Scalability:
- The system needs to be horizontally scalable to accommodate increasing user loads.
Reliability:
- The system should have a high level of availability; a good example is a 99.9% uptime rate.
- Regular data backups must be performed.
Security:
- User-provided data must be encrypted and kept in a secure environment.
- Reliable permission and authentication mechanisms are needed.
- Protecting against common internet vulnerabilities (such SQL injection and XSS) is essential.
Usability:
- The user interface must be straightforward and simple to use.
- Accessibility features for people with disabilities must be included.
Data Privacy:
- The system needs to comply with data protection regulations like the GDPR.
-User consent must be provided for data collection and use.
Data analytics:
- The program must provide information on user behaviour and movie reviews.
- These analytics can be applied to improve suggestions and get insights into the corporate environment.
Compatibility:
- The system must function with many web browsers and devices.
Maintenance:
- The code must be properly documented and follow industry standards.
- Regular maintenance and updates should be carried out.
Handling Load:
- The system must be able to handle peak loads during the release of new movies or other notable occurrences.
Cost- Effective:
- Cost-control measures need to be in place, particularly when it comes to hosting and server resources.
Regulatory Compliance:
- The system must adhere to all applicable movie content and rating laws.
Input and Monitoring:
- Implementation of technologies for monitoring system performance and gather user feedback for future improvements.
The Sentiment-Based Movie Rating System will certainly be able to satisfy user expectations while maintaining excellent standards of performance, security, and data privacy due to these functional and non-functional requirements.
5. Use Case
(Yifan Sheng s4584372)
Figure 5.1 Use Case Diagram
The use case diagram describes the operation of the website. The user should first be able to register and login to their own account. Then the user should be able to choose a particular movie from the category of movies which are stored in the websites database. Afterwards, the user can enter comments about the movie. The comments will be broken into words, and keywords will be recognized by GPT-3. It will detect and rank the sentimental level of these keywords. At last, the user will be able to view the comments from other users as well as the rating of this movie from GPT-3.
Use Case Name Register
Participating Actors New user
Flow of Events 1. The new user clicks on the register button in the homepage
2. The website forwards to the register page
3. The new user fills in all required section and click finish registration button, proceed to step 4
Or
The new user did not fill in all required sections or fill in with invalid data, error notification pop-up.
4. The new users file is sent and stored in the database
5. The website forwards to homepage while the new user is in log in status
Entry Condition The new user clicks on the register button in the homepage
Exit Condition The website forwards to homepage while the new user is in log in status
Quality Requirements The response time of the system as short as possible
Table 5.1 Use Case Register
Use Case Name Login
Participating Actors Registered user
Flow of Events 1. The registered user clicks on login button in the homepage
2. The website forwards to login page
3. The registered user enters username and password
4. The login request is sent to the server
5. If the login detail matches that users file stored in the database, the server approves the login request and the website forwards to homepage while the registered user is in log in status
Or
If the login detail does not match that users file stored in the database, the server denied the login request and the website stay in log in page with a pop-up notification of error message
Entry Condition The registered user clicks on login button in the homepage
Exit Condition The website forwards to the homepage while the registered user is in log in status.
Or
The website remains in login page with a pop-up notification of error message
Quality Requirements The response time of the system as short as possible
Table 5.2 Use Case Login
Use Case Name Search Movies
Participating Actors All user
Flow of Events 1. The user enters a Movie name in the search box in the homepage
2. The search request is sent to the server
3. If the Movie name is found in the database, the website will display the movies detail, comments, and rating
Or
If the Movie name is not found in the database, the website will display a message to tell the user that the movie is not found
Entry Condition The registered user clicks on the search button with valid content in the search box
Exit Condition The website displays the movies details, comments, and rating.
Or
The website displays a message to tell the user that the movie is not found
Quality Requirements The response time of the system as short as possible
Table 5.3 Use Case Search Movies
Use Case Name Enter comment
Participating Actors Logged in user
Flow of Events 1. A user logged in to the website and search for a movie exist in the database
2. The website displays the information of the movie as well as a text box for user to leave a comment
3. The user enters valid comment and click post button, proceed to step 4
Or
The user enters invalid comment and click post button, an error notification pop-up.
4. The comment is sent and store in the database
5. GPT-3 detects the sentiment of this comment and give rating to the movie based on it
6. The website displays a successful notification and remains in the homepage with movie information and the new comment added.
Entry Condition The logged in user enters valid comment and clicks the post button
Exit Condition The website displays a comment successful notification and remains in the homepage with movie information and the new comment added
Quality Requirements The response time of the system should be as short as possible
Table 5.4 Use Case Enter Comment
6. Sequence Diagrams
Adarsha bhandari(s4675648)
Register as a new user.
Figure 6.1: Register New User
Login
Figure 6.2: Login
Search for Movies
Figure 6.3 Search for Movies
Add comments on the Movies.
Figure 6.4: Adding Comments on the Movie
Edit Comments on the Movies.
Figure 6.5 Editing Comments on the Movie
7. Flow Chart
(Yifan Sheng s4584372)
Figure 7.1 Flow Chart
8. User Interface Design
(Yifan Sheng s4584372)
Figure 8.1 User Interface Design
9. Resource management plan
(Yifan Sheng s4584372)
Hardware: Computers and smartphones
Communication tools: WhatsApp and Zoom
Development tools: Microsoft Visio Studio, GitHub, Google doc, Google slides, Figma and Photoshop
Web server: Amazon EC2
Database server: Amazon RDS
Natural language processing system: GPT-3
Programming language:
HTML
CSS
JavaScript
PHP
Python
SQL
Communications between group members will be held on computers and smartphones via WhatsApp and Zoom. The development environment contains Microsoft Visio studio as main code editor and GitHub as a co-op coding and version control platform. The group will also use Google doc and Google slides as file management, as well as Figma for co-op user interface design. The back-end design will be based on AWS, which is Amazon Web Services. It provides not only a free 12-month trial of Amazon EC2 cloud server and Amazon RDS database, but also a thorough tutorial that helps the group to better manipulate the system. For the natural language processing system, OpenAI GPT-3 perfectly meets requirements of this project. As a powerful AI tool provided by OpenAI, GPT-3 has strong language comprehension ability, which can recognize and understand sentiments information in text, such as emotions, attitudes, tendencies, etc. Moreover, GPT-3 can undergo pre-training and fine-tuning to better adapt to our project requirements. Finally, it also provides free trials that cover the entire procedure of this project.
10. Risk management plan
(Yifan Sheng s4584372)
Risk description Likelihood Consequences Mitigation plan
User comments may contain illegal, hateful, or rude wording. Likely The experience of normal users may be affected, and it could also lead to website bans in certain countries and regions. Design filters to block inappropriate comments.
Some malicious users or bots may try to disrupt the system by entering meaningless sentences, garbled codes, or advertisements. Likely The rating system may capture keywords from those meaningless sentences, garbled codes, or advertisements. Design filters to filter out meaningful comments before capturing keywords for rating.
Users such as elderly movie enthusiasts who are not familiar with computers, may not know how to use rating systems. Likely The rating system may not be accurate enough due to insufficient samples. Make sure the website is user friendly when designing and add tutorials to the final product.
The system may not be able to break some long and difficult sentences or identify the keywords from them. Likely May lead to inaccurate ratings of movies on the website. Update the NLP model on time and use fine-tuning to train the model.
Users may accidentally misspell certain words or write sentences with language errors. High The rating system may not recognize these words or split these sentences. After reaching a certain number, it will affect the accuracy of the entire rating system. Adding an automatic spelling correction model during design to minimize situations where comments cannot be recognized.
The website may be threatened by malicious attacks. Likely The website may be paralysed, could lead to data loss, information hijacking, etc. Design sufficient network security protections, maintain the website in a timely manner, and fix vulnerabilities.
Table 10.1 Risk management table
11. Milestones
(Yifan Sheng s4584372)
Complete Project plan proposal by week 1 4/8/2023
Project Portfolio must be completed by week 3 (20/8/2023).
Complete oral presentation of the project portfolio by week 4 (August 23, 2023).
Complete the construction of the website, database, and system by the start of block 4.
Complete testing and beautification of the website by the end of Block 4.
Compared to refining the design progress every week, the group has decided to adopt a more flexible schedule. Because while implementing this project, group members also need to study other courses and complete corresponding assignments. Due to the different courses conducted by our group members, it is difficult for the group to set a synchronized and accurate timetable for the sentimental movie rating system in advance. Therefore, the group chose to only divide design steps into three: front-end website design, back-end server/database design, and testing/beautifying procedures. The group members will synchronize and advance the design progress through weekly supervisory meetings.
Figure 11.1 Development schedule
Figure 11.2 Development schedule in Gantt Chart
12. Conclusion
The Sentiment-based Movie Rating System aims to create a comprehensive online platform that will effectively curate and manage movie reviews while forecasting the corresponding ratings assigned to each critique. This digital nexus acts as a crucial medium for presenting an aggregated assessment, which makes up the comprehensive evaluation of a film. This continuum is ready to dynamically evolve as new viewer assessments seamlessly integrate. Our system's design distinguishes between two categories of users within this conceptual framework: registered and unregistered entities. The former enjoys a wide range of privileges, including access to movie trailers, the exposition of other viewers' perspectives, and a perceptive exposition of other viewers' emotional tonality, inferred through careful examination of their reviews. A multifaceted view of the cinematic landscape is simultaneously unfurled by the system, including genres, auteurs, writers, temporal spans, cinematic giants, and succinct narrative summaries. Together, these elements work to enhance the viewer's overall discernment. Importantly, the conversation goes beyond the surface, encompassing both laudatory praise and direct criticism, giving the audience a full picture of the discourse's depths. Additionally, the system coordinates the rise of the most acclaimed cinematic embodiments as a symbol of their artistic prowess.
The Sentiment Based Movie Rating System aspires to create a comprehensive online platform that will effectively curate and manage movie reviews while also forecasting the corresponding ratings assigned to each critique. A movie's holistic evaluation is presented using this digital nexus, which also acts as a key medium for presenting aggregated evaluations. This continuum is ready to dynamically change as new viewer evaluations are seamlessly integrated. This conceptual framework defines a distinction between registered and unregistered entities as user classifications for our system. The former enjoys a wide range of privileges, including access to movie trailers, the exposition of other viewers' perspectives, and a perceptive exposition of other viewers' emotional tonality, as inferred from a careful analysis of their reviews. The latter is also given access to movie trailers and other viewers' perspectives.
13. Contribution
Student Contribution Description
Yifan Sheng 33% Use Case
Flow Chart
User Interface Design
Resource management plan
Risk management plan
Milestones
Gantt chart
Pranav Dhakal 33% Project Background and Description
Project Scope
ConclusionReference
Adarsha Bhandari 33% Functional requirementsNonfunctional requirements
Sequence Diagrams
Related- Works
Table 13.1 Contribution table
15. References
Li, T., Zhong, S., & Xie, R. (2010). Privacy-Enhanced Data Publishing and Analysis. IEEE Transactions on Knowledge and Data Engineering, 22(3), 365-377. doi:10.1109/TKDE.2009.82
Alshammari, L., Grundy, J., & Hosking, J. (2017). Secure Software Development: A Systematic Literature Review. Information and Software Technology, 82, 131-153.
Asghari, H. J., Van Der Veen, M., & Herder, E. (2017). Understanding Online Privacy Policies: Labels and the Role of Audiences in Shaping Policy Requirements. Proceedings of the ACM on Human-Computer Interaction, 1(CSCW), 82. doi:10.1145/3134746
Shvetsova, Y. (2023, July 6). Functional and Non-Functional Requirements for Ecommerce Website | Elogic. Elogic. https://elogic.co/blog/functional-and-non-functional-requirements-for-ecommerce-websites/#:~:text=What%20are%20the%20functional%20requirements,process%2C%20social%20sharing%2C%20etc.
RaviK. et al. (2015) A survey on opinion mining and sentiment analysis: Tasks, approaches, and applications Knowledge-Based System
LiS. et al.(2015) Improving aspect extraction by augmenting a frequency-based method with web-based similarity measures Inf. Process. Manage.
BafnaK. et al. (2013) Feature based summarization of customers reviews of online products Procedia Comput. Sci.
DoH.H. et al. (2019) Deep learning for aspect-based sentiment analysis: A comparative review Expert Syst. Appl.
Amazon (2023). Amazon Web Services (AWS) - Cloud Computing Services. [online] Amazon Web Services, Inc. Available at: https://aws.amazon.com/.
openai.com. (n.d.). GPT-3 powers the next generation of apps. [online] Available at: https://openai.com/blog/gpt-3-apps.
