Search in social networks such as Facebook poses different challenges than in classical web search: besides the query text, it is important to take into account the searcher's context to provide relevant results. Their social graph is an integral part of this context and is a unique aspect of Facebook search. While embedding-based retrieval (EBR) has been applied in eb search engines for years, Facebook search was still mainly based on a Boolean matching model. In this paper, we discuss the techniques for applying EBR to a Facebook Search system. We introduce the unified embedding framework developed to model semantic embeddings for personalized search, and the system to serve embedding-based retrieval in a typical search system based on an inverted index. We discuss various tricks and experiences on end-to-end optimization of the whole system, including ANN parameter tuning and full-stack optimization. Finally, we present our progress on two selected advanced topics about modeling. We evaluated EBR on verticals 1 for Facebook Search with significant metrics gains observed in online A/B experiments. We believe this paper will provide useful insights and experiences to help people on developing embedding-based retrieval systems in search engines. CCS CONCEPTS • Information systems → Retrieval models and ranking; Search engine architectures and scalability; • Computing methodologies → Learning latent representations.

The Elias-Fano representation of monotone sequences has been recently applied to the compression of inverted indexes, showing excellent query performance thanks to its efficient random access and search operations. While its space occupancy is competitive with some state-of-the-art methods such as γ-δ-Golomb codes and PForDelta, it fails to exploit the local clustering that inverted lists usually exhibit, namely the presence of long subsequences of close identifiers.In this paper we describe a new representation based on partitioning the list into chunks and encoding both the chunks and their endpoints with Elias-Fano, hence forming a twolevel data structure. This partitioning enables the encoding to better adapt to the local statistics of the chunk, thus exploiting clustering and improving compression. We present two partition strategies, respectively with fixed and variablelength chunks. For the latter case we introduce a linear-time optimization algorithm which identifies the minimum-space partition up to an arbitrarily small approximation factor.We show that our partitioned Elias-Fano indexes offer significantly better compression than plain Elias-Fano, while preserving their query time efficiency. Furthermore, compared with other state-of-the-art compressed encodings, our indexes exhibit the best compression ratio/query time trade-off.

Entity linking deals with identifying entities from a knowledge base in a given piece of text and has become a fundamental building block for web search engines, enabling numerous downstream improvements from better document ranking to enhanced search results pages. A key problem in the context of web search queries is that this process needs to run under severe time constraints as it has to be performed before any actual retrieval takes place, typically within milliseconds.In this paper we propose a probabilistic model that leverages user-generated information on the web to link queries to entities in a knowledge base. There are three key ingredients that make the algorithm fast and space-efficient. First, the linking process ignores any dependencies between the different entity candidates, which allows for a O(k 2 ) implementation in the number of query terms. Second, we leverage hashing and compression techniques to reduce the memory footprint. Finally, to equip the algorithm with contextual knowledge without sacrificing speed, we factor the distance between distributional semantics of the query words and entities into the model.We show that our solution significantly outperforms several state-of-the-art baselines by more than 14% while being able to process queries in sub-millisecond times-at least two orders of magnitude faster than existing systems.

Graph reordering is a powerful technique to increase the locality of the representations of graphs, which can be helpful in several applications. We study how the technique can be used to improve compression of graphs and inverted indexes. We extend the recent theoretical model of Chierichetti et al. (KDD 2009) for graph compression, and show how it can be employed for compression-friendly reordering of social networks and web graphs and for assigning document identifiers in inverted indexes. We design and implement a novel theoretically sound reordering algorithm that is based on recursive graph bisection. Our experiments show a significant improvement of the compression rate of graph and indexes over existing heuristics. The new method is relatively simple and allows efficient parallel and distributed implementations, which is demonstrated on graphs with billions of vertices and hundreds of billions of edges