No content

Utilizing Embeddings 
 In Learning To Rank 
 For Search > Shawn TSAI / LINE Taiwan Data Dev 

Agenda > Search Everywhere > Search Result Relevance > Embeddings > Learning To Rank > Search Workflow 

Search Everywhere Life on LINE 

Search Result Relevance > The main goal is to reduce the semantic gap between user query and documents. > The key points: semantic features and ranking function. > Search is a ranking problem. The ordering is more important than the predicted probability of a single instance. 

> 聊天記錄⼀一直在資料壓縮中 > 不管怎麼按備份聊天紀錄都不能備份 Limitation: Different description Limitation: No shared keywords > 為什什麼有時候賴都不會通知 > 訊息都跑不出來來是怎樣 Search Scoring & Limitation > , > ( ) = () = 1 + + 1 + 1 _ = ∗ Standard similarity function: TF-IDF 

Embeddings 

Word Embedding > Vector representation > Capturing context of a word in a document, semantic/syntactic similarity, relation with other words Source: Efficient Estimation of Word Representations in Vector Space 

BERT BERT is a new method of pre-training language representations which obtains state-of-the-art results on a wide array of NLP tasks. Source: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding Bidirectional Encoder Representations from Transformers 

Querying By Vector Representation Sent. Encoding
 By Pre-trained BERT
 Model Query Document Vecs Index Query Vec Documents Document Vecs Online Offline Nearest Neighbor Search Build 
 NN 
 Index 

Learning To Rank 

Learning To Rank > Applying machine learning to construct ranking models for information retrieval systems > Caring more about ranking rather than rating prediction > Scoring by machine learning • Creating document index by Elasticsearch • Using embeddings to train ranking models • Serving search queries by Elasticsearch with ranking models 

Filters Search Architecture Documents Query Filter Index ES + Re-ranking BERT Matches Ranked Results NER … Scoring Index Ranking Models 

Ƃ Custom Scoring Function Ƃ 

Search Workflow With Learning To Rank User’s Needs Measure Relevance Pre-process Inverted-index Features Selection Ranking Models Scoring Function NDCG MAP Precision@k Deploy Monitoring Feedback Evaluation Build Index Learning To Rank Serve Data 

More Consideration > Good judge lists matching user needs of search quality > Good metrics measuring search results > Incorporating with embeddings into scoring function > Synchronizing the version between indexing and serving layers > A/B testing 

Thank you