Facebook, Inc.

Madhu Ramanathan

Juan Miguel Pino

Chun-Yang Chen

Zhongxian Chen

Taroon Mandhana

Embodiments are disclosed for transliteration based on a machine translation model training pipeline. A method according to some embodiments includes steps of: training a probabilistic model for transliteration from a first script system to a second script system using a machine translation model training pipeline; segmenting, using the probabilistic model, an input string in the first script system into phonemes that correspond to characters in the second script system; converting the phonemes in the first script system into the characters in the second script system, the characters forming a word or a word prefix in the second script system; and outputting the word or the word prefix in the second script system.

Transliteration using machine translation pipeline

In one embodiment, an online social network may receive, from a client system of a first user of the online social network, a search query inputted by the first user; and identify one or more of the entities matching the search query. At least a portion of the entities being identified based at least in part on a recorded geographic proximity between the first user and each identified entity. The online social network may access one or more proximity coefficients for the identified entities from a proximity-coefficient system of the online social network; calculate a score for each identified entity based at least in part on the value of the accessed proximity coefficients for the identified entity and social-graph information associated with the identified entity; and send, to the client system of the first user in response to the received search query, search results for display to the first user.

Growing community of inventors

Mountain View, CA, United States of America

Madhu Ramanathan