A Fast Deep Learning Model for Textual Relevance in Biomedical Information Retrieval

Mohan, Sanjay; Fiorini, Nicolas; Kim, Sun; Lu, Zhiyong

doi:10.1145/3178876.3186049

Cited by 33 publications

(18 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yang et al [5] proposed the concept of trust circle in the social network, trying to ease the pressure of user data sparseness through the friends historical preferences as the user's interest feature should be similar to the average of his friends characteristics. As deep learning has made breakthroughs in multiple research areas such as computer vision [1], natural language processing [16],information retrieval [17] and so on. It has also been gradually applied to the RS.…”

Section: Sparse and Cold Startmentioning

confidence: 99%

TDCTFIC: A Novel Recommendation Framework Fusing Temporal Dynamics, CNN-Based Text Features and Item Correlation

Xiong

Feng

et al. 2019

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

The traditional recommendation system (RS) can learn the potential personal preferences of users and potential attribute characteristics of items through the rating records between users and items to make recommendations.However, for the new items with no historical rating records,the traditional RS usually suffers from the typical cold start problem. Additional auxiliary information has usually been used in the item cold start recommendation,we further bring temporal dynamics,text and relevance in our models to release item cold start.Two new cold start recommendation models TmTx(Time,Text) and TmTI(Time,Text,Item correlation) proposed to solve the item cold start problem for different cold start scenarios.While well-known methods like TimeSVD++ and CoFactor partially take temporal dynamics,comments,and item correlations into consideration to solve the cold start problem but none of them combines these information together.Two models proposed in this paper fused features such as time,text,and relevance can effectively improve the performance under item cold start.We select the convolutional neural network (CNN) to extract features from item description text which provides the model the ability to deal with cold start items.Both proposed models can effectively improve the performance with item cold start.Experimental results on three real-world data set show that our proposed models lead to significant improvement compared with the baseline methods.

show abstract

Section: Sparse and Cold Startmentioning

confidence: 99%

TDCTFIC: A Novel Recommendation Framework Fusing Temporal Dynamics, CNN-Based Text Features and Item Correlation

Xiong

Feng

et al. 2019

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Best Match [6] is the first relevance search algorithm for PubMed that leverages the intelligence of users and machine-learning technology as an alternative to the traditional sorting techniques. Delta [25] is a deep learning based model that applies convolution operation upon an updated document matrix in which each word is replaced with the most similar word in the query. However, this model takes the key topic word and other words in query with equal importance, making the retrieval out of focus.…”

Section: Related Work 21 Biomedical Literature Retrievalmentioning

confidence: 99%

“…One is the representation-focused model, which tries to learn good representations for both query and documents with deep neural networks, and then conducts matching between the learned representations. Examples include DSSM [16], C-DSSM [9], ARC-I [15], Delta [25], MASH RNN [19]. The other is the interaction-focused model, which first builds local interactions (i.e., local matching signals) between the query and documents, and then uses deep neural networks to learn the overall matching score.…”

Section: Document Retrieval Modelsmentioning

confidence: 99%

“…Although this increases recall, it often decreases precision which is crucial for precision medicine [13]. The second type is deep learning models, such as DeepMatch [22] and Delta [25], which have demonstrated superior performance in recent years. The advantage of these models comes from their flexible representation learning process for the query and the documents.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Graphene

Zhao

Sboner

et al. 2019

Proceedings of the 28th ACM International Conference on Information and Knowledge Management

View full text Add to dashboard Cite

Effective biomedical literature retrieval (BLR) plays a central role in precision medicine informatics. In this paper, we propose GRAPHENE, which is a deep learning based framework for precise BLR. GRAPHENE consists of three main different modules 1) graph-augmented document representation learning; 2) query expansion and representation learning and 3) learning to rank biomedical articles. The graph-augmented document representation learning module constructs a document-concept graph containing biomedical concept nodes and document nodes so that global biomedical related concept from external knowledge source can be captured, which is further connected to a BiLSTM so both local and global topics can be explored. Query expansion and representation learning module expands the query with abbreviations and different names, and then builds a CNN-based model to convolve the expanded query and obtain a vector representation for each query. Learning to rank minimizes a ranking loss between biomedical articles with the query to learn the retrieval function. Experimental results on applying our system to TREC Precision Medicine track data are provided to demonstrate its effectiveness.

show abstract

“…We use the sets from (Park and Chiba, 2017) available online, enriched with user and time information provided in the original AOL dataset. In addition, we evaluate the systems on a second real-world dataset from a production search engine in the biomedical domain, PubMed (Fiorini et al, 2017;Lu, 2011;Mohan et al, 2018), that was created in the same manner. The biomedical dataset consists of 8,490,317 queries.…”

Section: Datasetmentioning

confidence: 99%

Personalized neural language models for real-world query auto completion

Fiorini¹,

Lu²

2018

Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Hu

Self Cite

View full text Add to dashboard Cite

Query auto completion (QAC) systems are a standard part of search engines in industry, helping users formulate their query. Such systems update their suggestions after the user types each character, predicting the user's intent using various signals -one of the most common being popularity. Recently, deep learning approaches have been proposed for the QAC task, to specifically address the main limitation of previous popularity-based methods: the inability to predict unseen queries. In this work we improve previous methods based on neural language modeling, with the goal of building an end-to-end system. We particularly focus on using real-world data by integrating user information for personalized suggestions when possible. We also make use of time information and study how to increase diversity in the suggestions while studying the impact on scalability. Our empirical results demonstrate a marked improvement on two separate datasets over previous best methods in both accuracy and scalability, making a step towards neural query auto-completion in production search engines.

show abstract

A Fast Deep Learning Model for Textual Relevance in Biomedical Information Retrieval

Cited by 33 publications

References 26 publications

TDCTFIC: A Novel Recommendation Framework Fusing Temporal Dynamics, CNN-Based Text Features and Item Correlation

TDCTFIC: A Novel Recommendation Framework Fusing Temporal Dynamics, CNN-Based Text Features and Item Correlation

Graphene

Personalized neural language models for real-world query auto completion

Contact Info

Product

Resources

About