H.3.8. Natural Language Processing
Mozhgan Akaberi; Maryam Khodabakhsh; Seyedehfatemeh Karimi; Hoda Mashayekhi
Abstract
The exponential growth of digital information has increased the demand for robust and efficient Information Retrieval (IR) systems. Query Performance Prediction (QPP) is a critical task for identifying difficult queries and enhancing retrieval strategies. However, existing QPP methods suffer from several ...
Read More
The exponential growth of digital information has increased the demand for robust and efficient Information Retrieval (IR) systems. Query Performance Prediction (QPP) is a critical task for identifying difficult queries and enhancing retrieval strategies. However, existing QPP methods suffer from several limitations: (1) score-based approaches fail to capture the structural relationships among retrieved documents, (2) supervised methods require labeled training data, making them costly and impractical for new domains, and (3) unsupervised post-retrieval predictors often rely solely on retrieval score dispersion, neglecting document clustering effects. To address these challenges, we propose a novel clustering-based post-retrieval QPP method. Specifically, we introduce three unsupervised predictors: Clustered Distinction, which measures query-specific separability of retrieved clusters; Clustered Query Drift, which estimates the deviation of top-ranked documents from query intent; and a hybrid approach combining both. By analyzing the clustering structure of retrieved documents, our method improves interpretability while eliminating the need for labeled data. We evaluate our approach on three standard datasets: the large-scale MS MARCO Passage Ranking dataset, TREC DL 2019, and TREC DL 2020. Experimental results demonstrate that our method significantly outperforms state-of-the-art score-based QPP models. These findings highlight the potential of cluster-aware QPP for enhancing IR systems and reducing the impact of difficult queries.
H.3.2.10. Medicine and science
Fahimeh Hafezi; Maryam Khodabakhsh
Abstract
Coronavirus disease as a persistent epidemic of acute respiratory syndrome posed a challenge to global healthcare systems. Many people have been forced to stay in their homes due to unprecedented quarantine practices around the world. Since most people used social media during the Coronavirus epidemic, ...
Read More
Coronavirus disease as a persistent epidemic of acute respiratory syndrome posed a challenge to global healthcare systems. Many people have been forced to stay in their homes due to unprecedented quarantine practices around the world. Since most people used social media during the Coronavirus epidemic, analyzing the user-generated social content can provide new insights and be a clue to track changes and their occurrence over time. An active area in this space is the prediction of new infected cases from Coronavirus-generated social content. Identifying the social content that relates to Coronavirus is a challenging task because a significant number of posts contain Coronavirus-related content but do not include hashtags or Corona-related words. Conversely, posts that have the hashtag or the word Corona but are not really related to the meaning of Coronavirus and are mostly promotional. In this paper, we propose a semantic approach based on word embedding techniques to model Corona and then introduce a new feature namely semantic similarity to measure the similarity of a given post to Corona in semantic space. Furthermore, we propose two other features namely fear emotion and hope feeling to identify the Coronavirus-related posts. These features are used as statistical indicators in a regression model to estimate the new infected cases. We evaluate our features on the Persian dataset of Instagram posts, which was collected in the first wave of Coronavirus, and demonstrate that the consideration of the proposed features will lead to improved performance of the Coronavirus incidence rate estimation.
