KindleMeSome is a search engine project designed to process and retrieve book and review information from Amazon's Kindle store. The project leverages the Apache Solr platform to implement a scalable and efficient search engine. The data, sourced from Amazon's Kindle store, includes detailed metadata about books and associated user reviews. The primary focus of the project was to create a system capable of handling complex queries, providing accurate and relevant search results, and evaluating the performance of these queries using standard information retrieval metrics.
Read more about it in the project report or check the presentation slides for a quick overview.
- Data Processing and Exploration pipeline.
- Data visualization and analysis using Python libraries (e.g., Matplotlib, Seaborn).
- Apache Solr schemas for reviews and books.
- Indexing and Querying with Apache Solr specific configurations.
- Searching for books by specific authors.
- Filtering books based on content length and topic.
- Retrieving reviews based on specific sentiments (e.g., positive reviews, vacation reading recommendations).
- Faceted search for filtering search results.
- Cursor-based pagination for handling large result sets.
- Suggestion for auto-completion of search queries (Fuzzy search).
- Highlighting of search terms in the search results.
- Performance evaluation using standard information retrieval metrics (e.g., Mean Average Precision, Recall, Precision).
- React frontend for user interaction with the search engine.
| Name | Number | GitHub | |
|---|---|---|---|
| António Ribeiro | up201906761 | up201906761@edu.fe.up.pt | @francisco-rente |
| Diogo Maia | up201904974 | up201904974@edu.fe.up.pt | @DFMM11 |
| Luís Viegas | up201904979 | up201904979@edu.fe.up.pt | @luis-viegas |
Data was gathered from two different datasets:
- Kaggle's Amazon Kindle book reviews
- UCSD Amazon review data from 2018 (Kindle Store's metadata section)
⚠️ Datasets: The second link that stored the dataset is no longer available. The pipeline stage is not reproducible, and consequently, the final version of the project is not currently available as a full demonstration.
