Academic Year 2023 - 24
S8 PROJECT WORK II
Third Review
PROJECT TITLE
MEDIA RECAP COMPANION
BIP PROJECT ID CATERGORY
24S8EOW436 EXTERNAL – OWN
BATCH MEMBERS GUIDE
SOWMIYA S(202IT224) Prof. NIKITHA M
SARAVANAN T(202IT213) Assistant Professor
BHARANIDHARAN M(202IT119) Department of Information Technology
LITERATURE SURVEY
Works carried out
( with details of Methods/ Information
Sl.No. Journal Paper Title with Author Materials/ Software/ gathered relevant to
Algorithms / techniques/ your project
components used )
K. Chen et al., "Extractive Broadcast News Methods, Algorithms Neural Network
Summarization Leveraging Recurrent Language Modeling
Neural Network Language Modeling (RNNLM) Framework
1 Techniques," in IEEE/ACM Transactions on
Audio, Speech, and Language Processing,
vol. 23, no. 8, pp. 1322-1334, Aug. 2015, doi:
10.1109/TASLP.2015.2432578
M. Steinert, R. Granada, J. P. Aires and F. Algorithms LexRank, TextRank,
Meneguzzi, "Automating News Latent Semantic
Summarization with Sentence Vectors Analysis (LSA), T5,
2 Offset," 2019 8th Brazilian Conference on BART and PEGASUS
Intelligent Systems (BRACIS), 2019, pp.
102-107, doi: 10.1109/BRACIS.2019.00027
25-03-2024 THIRD REVIEW PRESENTATION 2
LITERATURE SURVEY
Works carried out
( with details of
References Methods/ Materials/ Information gathered
Sl.No. ( Journal Papers Only) Software/ Algorithms / relevant to your
techniques/ project
components used )
Rahul, S. Adhikari and Monika, "NLP based Algorithms Text Rank Algorithm,
Machine Learning Approaches for Text Latent Semantic analysis
Summarization," 2020 Fourth International Algorithm
3 Conference on Computing Methodologies
and Communication (ICCMC), 2020, pp.
535-538.
S. Singh, A. Singh, S. Majumder, A. Sawhney, Methods, Algorithms BART, Text
D. Krishnan and S. Deshmukh, "Extractive summarization , NLP,
Text Summarization Techniques of News statistical analysis
4 Articles: Issues, Challenges and methods.
Approaches," 2019 International
Conference on Vision Towards Emerging
Trends in Communication and Networking
25-03-2024 THIRD REVIEW PRESENTATION 3
AIM & OBJECTIVES OF THE PROJECT
(MEDIA RECAP COMPANION)
AIM:
To develop A web based application chrome extension for global to
enhance with the daily news.
OBJECTIVE:
• Automated Summarization
• Personalized Content Aggregation
• Multi-format Recap
• Real-time Updates And Alerts
25-03-2024 THIRD REVIEW PRESENTATION 4
SCOPE OF THE PROJECT
• Diverse News Sources: Our project scope includes aggregating news from a variety of
reputable sources, ensuring our listeners receive a well-rounded perspective on current
events, fostering a comprehensive understanding.
• Customized Content Delivery: With a focus on user preferences, our project aims to offer
customizable content delivery, allowing listeners to tailor their news summaries based on
their interests, creating a personalized and engaging experience.
• Real-Time Updates: The scope involves providing real-time news updates, ensuring our
listeners are promptly informed about the latest developments, contributing to a dynamic and
relevant news consumption experience.
• Multilingual Accessibility: We aspire to make our project accessible to a global audience by
incorporating multilingual features, allowing listeners to enjoy news summaries in their
preferred language and promoting inclusivity in information consumption.
25-03-2024 THIRD REVIEW PRESENTATION 5
NEED FOR THE CURRENT STUDY
• Information Explosion: News summarization helps cut through the noise and provides
concise insights.
• AI and NLP Advancements: Advances in Artificial Intelligence (AI) and Natural Language
Processing (NLP) have led to more sophisticated summarization techniques. Ongoing research
explores how these technologies can improve the accuracy and coherence of news summaries.
• Fake News Mitigation: News summarization, when combined with fact-checking
mechanisms, can contribute to the identification and clarification of accurate information.
• Personalization and Recommendation Systems: News summarization supports the
customization of content based on user preferences, leading to more relevant
recommendations and increased user satisfaction.
25-03-2024 THIRD REVIEW PRESENTATION 6
FEASIBILITY ANALYSIS -1
TECHNICAL FEASIBILITY
• Technology Stack: The chosen technologies (Streamlit, BeautifulSoup) are suitable for the
project.
• Data Processing: Evaluate the efficiency of data processing, especially when dealing with a
large number of news articles.
FUNCTIONAL FEASIBILITY
• Features: Ensure that the application meets its functional requirements and provides
value to users.
• Integration: Check if the application can be easily integrated into existing systems or
workflows.
SCALABILITY
• Performance: Assess how well the application performs as the number of users or news
articles increases.
• Scalability: Consider whether the architecture allows for easy scaling if the user base
grows.
25-03-2024 THIRD REVIEW PRESENTATION 7
FEASIBILITY ANALYSIS -2
OPERATIONAL FEASIBILITY:
• User Interface: Assess the ease of use of the system. The user interface should be intuitive
for users to interact with and should not require extensive training.
• Integration: Consider how easily the summarization system can integrate with existing
newspaper systems or platforms.
USER FEEDBACK:
• User Testing: Collect feedback from potential users through testing to identify any
usability issues.
• User Expectations: Ensure that the application aligns with user expectations for a news
summarizer.
25-03-2024 THIRD REVIEW PRESENTATION 8
CHOICE OF COMPONENTS & MODULES
TEXT SUMMARIZATION MODULE
• Utilize natural language processing (NLP) techniques and libraries (such as NLTK or
SpaCy) for extracting key information and generating concise text summaries.
CONTENT AGGREGATION AND SCRAPING MODULE
• We design a web scraping or API-based module to fetch content from various media
sources.
• Apply content aggregation techniques to compile diverse information related to users'
preferences.
REAL-TIME UPDATES AND ALERTS MODULE
• We use asynchronous programming for real-time updates and push notifications.
• Integrate with news APIs or feeds to fetch breaking news.
USER INTERFACE (UI) DESIGN MODULE
• We choose a front-end framework in python for building an interactive and user-friendly
interface.
• We implement responsive design for seamless user experience across devices.
25-03-2024 THIRD REVIEW PRESENTATION 9
CHOICE OF METHODS & TECHNIQUES
MACHINE LEARNING FOR SUMMARIZATION
• Explore machine learning models (LexRank, TextRank, Latent Semantic Analysis (LSA), T5,
BART and PEGASUS) for more advanced and context-aware summarization.
PERSONALIZATION ALGORITHMS
• Implement recommendation algorithms to enhance personalization based on user
behavior and preferences.
CONTINUOUS INTEGRATION/CONTINUOUS DEPLOYMENT (CI/CD)
• Employ CI/CD pipelines for automated testing and deployment, ensuring a smooth
development process.
25-03-2024 THIRD REVIEW PRESENTATION 10
CHOICE OF EQUIPMENT USED FOR PROJECT DEVELOPMENT
DEVELOPMENT ENVIRONMENT
• Use integrated development environments (IDEs) such as Visual Studio Code for coding.
VERSION CONTROL
• Employ version control systems Git for collaborative development.
TESTING FRAMEWORKS
• Integrate testing frameworks for automated testing of different modules.
25-03-2024 THIRD REVIEW PRESENTATION 11
PROJECT WORK PLAN
Project Duration: Estimated 3 - 4 months
Phase -1 : Project Initiation (2 Weeks)
1. Project Scope
2. Resource Planning
3. Technology Stack Selection
Phase -2 : Data Collection and Preparation (3-4 weeks)
1. Data – Article Gathering
2. Data Preprocessing
Phase -3 : Model Development (3-4 weeks)
1. Model Architecture
2. Training
25-03-2024 THIRD REVIEW PRESENTATION 12
PROJECT WORK PLAN
Phase -4 : Evaluation and Testing (1-2 weeks)
1. Model Evaluation
2. Testing and Validation
Phase -5 : Deployment (1-2 weeks)
1. Deployment Strategy
2. User Interface
Phase 6: Maintenance and Continuous Improvement (Ongoing)
1. Monitoring and Maintenance
2. Feedback Loop
25-03-2024 THIRD REVIEW PRESENTATION 13
PROJECT WORK PLAN
Phase 7: Documentation and Knowledge Transfer(Throughout the Project)
1. Documentation
2. Knowledge Transfer
Phase 8: Reporting and Conclusion (Final Month)
1. Final Report
2. Conclusion
25-03-2024 THIRD REVIEW PRESENTATION 14
EFFECTIVE UTILIZATION OF THE MODERN TOOL
STREAMLIT:
• User-Friendly Interface
• Interactive Visualizations
• Real-Time Updates
GOOGLE API:
• Advanced Search and Filtering
• Reliability and Scalability
• Comprehensive Coverage
25-03-2024 THIRD REVIEW PRESENTATION 15
PROPOSED METHODOLOGY ( Flow Chart)
25-03-2024 THIRD REVIEW PRESENTATION 16
PROPOSED METHODOLOGY ( Gantt Chart)
WEEK 1 WEEK 2 WEEK 3 WEEK 4 WEEK 5 WEEK 6 WEEK 7 WEEK 8
Learn
Literature
Survey
Content
Study on
Algorithm
Working on
frontend
Working on
Backend
Testing and
Debugging
Update
25-03-2024 THIRD REVIEW PRESENTATION 17
MODULES
TEXT SUMMARIZATION MODULE
• Utilize natural language processing (NLP) techniques and libraries (such as NLTK or
SpaCy) for extracting key information and generating concise text summaries.
CONTENT AGGREGATION AND SCRAPING MODULE
• We design a web scraping or API-based module to fetch content from various media
sources.
• Apply content aggregation techniques to compile diverse information related to users'
preferences.
REAL-TIME UPDATES AND ALERTS MODULE
• We use asynchronous programming for real-time updates and push notifications.
• Integrate with news APIs or feeds to fetch breaking news.
USER INTERFACE (UI) DESIGN MODULE
• We choose a front-end framework in python for building an interactive and user-friendly
interface.
• We implement responsive design for seamless user experience across devices.
25-03-2024 THIRD REVIEW PRESENTATION 18
RESULT AND DISCUSSION
• The primary goal of news summarization is to distill the main points, essential details, and
crucial context of a news story while omitting less significant information
• Demo in live
25-03-2024 THIRD REVIEW PRESENTATION 19
CONCLUSION
• Distill the main points, essential details, and crucial context of a news story while
omitting less significant information.
• Summarization of news serves as a vital tool in managing the information overload
in the digital age.
• Extractive or Abstractive methods in summarization techniques contribute to a
more efficient and accessible way of staying informed about current events.
• This process requires sophisticated natural language processing (NLP) techniques
and machine learning algorithms to understand the content, identify important
elements, and generate coherent summaries.
• Continued advancements in NLP and machine learning will likely further refine and
enhance the capabilities of news summarization systems in the future.
25-03-2024 THIRD REVIEW PRESENTATION 20
INDIVIDUAL CONTRIBUTIONS TO THE WORK
Batch Member 1 : (202IT224-SOWMIYA S )
1. Coordinate the team members.
2. News Article collection related to work.
3. Study the research paper to work on the negative area.
4. Lists the important methods and models in research Paper.
5. To upload the weekly report , surveys and get guidelines from Guide to
maintain the project ongoing.
6. To work on writing the project report paper.
25-03-2024 THIRD REVIEW PRESENTATION 21
INDIVIDUAL CONTRIBUTIONS TO THE WORK
Batch Member 2 : (202IT213 – SARAVANAN T)
1. To write the program according to the list of models selected.
2. Helps in choosing the research papers.
3. To work on writing literature survey according the team work separation.
4. To update the weekly review to the guide and get changes in code and
update in project ideas.
5. To maintain a GitHub repository up to date for research purpose.
25-03-2024 THIRD REVIEW PRESENTATION 22
INDIVIDUAL CONTRIBUTIONS TO THE WORK
Batch Member 3 : (202IT158 –BHARANIDHARAN M)
1. To find out the research paper related to our project.
2. To speed up the model program work by collaborating the process of
programming work.
3. To find out the best competition site , publications site to upload our process.
25-03-2024 THIRD REVIEW PRESENTATION 23
Completion and Submission of Project Report
List of Documents to be Submitted
SL.No List of Documents Status ( Provide the drive link
of prepared document)
1
Cover Page & Title Page (Both are in cover page
same format)
2 Bonafide Certificate Bonafide certificate
3 Declaration Declaration
4 Acknowledgement Acknowlegement
5 Chapter I – Introduction Chapter I - Introduction
6 Chapter 2 – Literature Survey Chapter 2 - Literature Survey
25-03-2024 THIRD REVIEW PRESENTATION 24
Submission of Project Report
List of Documents to be Submitted
SL.No List of Documents Status ( Provide the drive link
of prepared document)
7 Chapters 3 - Objectives and Chapter 3 - Objectives
Methodology
8 Chapters 4 - Proposed work modules Chapter 4 -Proposed Work
9 Chapter 5 - Results and Discussion Chapter 5 - Results and Discus
sion
10 Chapters 6 - Conclusions & Suggestions Chapter 6 - Conclusions
for Future work
11 References References
12 Appendices
25-03-2024 THIRD REVIEW PRESENTATION 25
PUBLICATIONS (CONFERENCE / JOURNAL / PATENT)
SL.No Paper/Manuscript Title Status ( Provide the drive link
of prepared document)
1
MEDIA RECAP COMPANION Paper work
25-03-2024 THIRD REVIEW PRESENTATION 26
REFERENCES
( Journal Papers/ Books/ Website References )
• J. YaA. Nenkova and K. McKeown, "Automatic summarization", Foundat. Trends Inf. Retrieval, vol. 5, no. 2–3, pp. 103-233, 2011.
• I. Mani and M. T. Maybury, Advances in Automatic Text Summarization, USA, MA, Cambridge:MIT Press, 1999.
• Y. Gong and X. Liu, "Generic text summarization using relevance measureand latent semantic analysis", Proc. ACM SIGIR Conf. Res.
Develop. Inf. Retrieval, pp. 19-25, 2001.
• G. Erkan and D. R. Radev, "LexRank: Graph-based lexical centrality as saliencein text summarization", J. Artif. Intell. Res., vol. 22, no. 1,
pp. 457-479, 2004.
• Nallapati R, Zhou B, Santos C, Gulcehre, Xiang B (2016) Abstractive text summarization using sequence-to-sequence RNNs and
beyond. In: Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning, pp 280–290
• Tianyi Zhang*, Varsha Kishore*, Felix Wu*, Kilian Q. Weinberger, and Yoav Artzi. 2020. BERTScore: Evaluating Text Generation with
BERT.
• Jingqing Zhang, Yao Zhao, Mohammad Saleh, and Peter Liu. 2020. PEGASUS: Pre-training with extracted gap-sentences for abstractive
summarization. In International Conference on Machine Learning, pages 11328–11339. PMLR.
• Karen Spä rck Jones. 2007. Automatic summarising: The state of the art. Information Processing & Management, 43(6):1449–1481.
• NewsAPI,”A Web Service for Accessing News Articles Programmatically,” [Online]. Available: https://newsapi.org/.
25-03-2024 THIRD REVIEW PRESENTATION 27
