# US 2014/0156565 A1 · Methods and systems for predicting learning curve for statistical machine translation system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0156565 A1 - Document ID: US20140156565A1 - Status: Published application - Kind: A1 - Application number: 13/689,878 - Filed: 2012-11-30 - Published: 2014-06-05 - Earliest claimed: 2012-11-30 - Assignees: Xerox Corporation - Applicants: Xerox Corporation - Inventors: Prasanth Kolachina; Nicola Cancedda; Marc Dymetman; Sriram Venkatapathy - Web page: https://patentgrep.com/patent/US20140156565A1 ## CPC - G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} - G06F 40/51: Translation evaluation - G06F 40/44: Statistical methods, e.g. probability models - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract The disclosed embodiments relate to a system and method for predicting the learning curve of an SMT system. A set of anchor points are selected. The set of anchor points correspond to a size of a corpus. Thereafter, a gold curve or a benchmark curve is fitted based on the set of anchor points to determine the BLEU score. Based on the BLEU score and a set of parameters associated with the first set of anchor points, a confidence score is computed. ## Claims ### Claim 1 A computer implemented method of training a prediction model to estimate a learning curve indicative of training of a statistical machine translation (SMT), the computer implemented method comprising: selecting a first set of anchor points, wherein each of the first set of anchor points corresponds to a different number of monolingual segments, wherein the first set of anchor points are selectable by a user; computing a weight vector for each of the first set of anchor points based, at least in part, on a first set of parameters associated with each of the first set of anchor points; and computing a first confidence score for each of the first set of anchor points based on the weight vector. ### Claim 2 (depends on 1) The computer implemented method of claim 1 further comprising computing a second set of parameters associated with the learning curve based, at least in part, on the weight vector and the first confidence score. ### Claim 3 (depends on 1) The computer implemented method of claim 1, wherein the computing of the first confidence score comprises: fitting a first curve based on each of the first set of anchor points to determine a first score for each of the first set of anchor points; computing the weight vector for each of the first set of anchor points based, at least in part, on a regression model and the first set of parameters, and the first score and computing the first confidence score for each of the first set of anchor points based on the weight vector, the first set of parameters, and the first score. ### Claim 4 (depends on 1) The computer implemented method of claim 1, wherein the first set of parameters include average length of a segment, average length of tokens, lexical diversity, word-order divergence. ### Claim 5 (depends on 1) The computer implemented method of claim 1, wherein the second set of parameters includes a value at an origin, a slope at the origin, and an asymptotic value. ### Claim 6 (depends on 1) The computer implemented method of claim 1, wherein the first set of anchor points includes at least three anchor points. ### Claim 7 (depends on 1) The computer implemented method of claim 1 further comprising displaying the learning curve to the user. ### Claim 8 A computer implemented method of estimating a learning curve indicative of training of a statistical machine translation (SMT), the computer implemented method comprising: selecting a second set of anchor points, wherein each of the second set of anchor points corresponds to a different number of bi-lingual segments, wherein the second set of anchor points are selectable by a user; fitting a second curve based on each of the second set of anchor points; and estimating a second set of parameters of the second curve to obtain the learning curve, wherein the second set of parameters include a value at an origin, a slope at the origin, and an asymptotic value. ### Claim 9 (depends on 8) The computer implemented method of claim 8, wherein the estimating corresponds to at least one of extrapolating or interpolating. ### Claim 10 (depends on 8) The computer implemented method of claim 8 further comprising fitting a first curve based on the second set of anchor points. ### Claim 11 (depends on 10) The computer implemented method of claim 10 further comprising computing a root mean square error based on the learning curve and the first curve. ### Claim 12 (depends on 8) The computer implemented method of claim 8 further comprising displaying the learning curve to the user. ### Claim 13 A computer implemented method of training a prediction model to estimate a learning curve associated with training of a statistical machine translation (SMT), the computer implemented method comprising: selecting a first set of anchor points, wherein each of the first set of anchor points corresponds to a different number of monolingual segments, wherein the first set of anchor point are selectable by a user; selecting a second set of anchor points, wherein each of the second set of anchor points corresponds to a different number of bi-lingual segments, wherein the second set of anchor point are selectable by the user; fitting a second curve based on the second set of anchor points; extending the second curve to the first set of anchor points; computing a second confidence score for each of the first set of anchor points based, at least in part, on one or more parameters associated with the second curve; computing a weight vector for each of the first set of anchor points based on a first set of parameters associated with each of the first set of anchor points; and computing a first confidence score for each of the first set of anchor points based on the weight vector. ### Claim 14 (depends on 13) The computer implemented method of claim 13 further comprising computing a second set of parameters associated with the learning curve based, at least in part on, the first confidence score, the second confidence score, and the weight vector. ### Claim 15 (depends on 13) The computer implemented method of claim 13, wherein the computing the first confidence score further comprises: fitting a first curve based on each of the first set of anchor points to determine a first score for each of the first set of anchor points; computing the weight vector for each of the first set of anchor points based, at least in part, on a regression model, the first set of parameters, and the first score; and computing the first confidence score for each of the first set of anchor points based on the weight vector, the first set of parameters and the first score. ### Claim 16 (depends on 15) The computer implemented method of claim 15, wherein the first curve corresponds to a benchmark curve. ### Claim 17 (depends on 13) The computer implemented method of claim 13, wherein the first set of parameters include average length of a segment, average length of tokens, lexical diversity, word-order divergence. ### Claim 18 (depends on 13) The computer implemented method of claim 13, wherein the second set of parameters includes a slope, asymptotic value and steady state point. ### Claim 19 (depends on 13) The computer implemented method of claim 13, wherein extending corresponds to at least one of extrapolating or interpolating. ### Claim 20 (depends on 13) The computer implemented method of claim 13 further comprising displaying the learning curve to the user. ### Claim 21 A prediction model server for estimating a learning curve indicative of training of a SMT, the system comprising: an anchor point selection module configured to select at least one of a first set of anchor points or a second set of anchor points; a curve fitting module configured to fit a second curve based on each of the second set of anchor points; an estimation module configured to extrapolate the second curve to the first set of anchor points; a scoring module configured to: compute a second confidence score for each of the first set of anchor points based, at least in part, on one or more parameters associated with the second curve; and compute a weight vector for each of the first set of anchor points based on a first set of parameters associated with each of the first set of anchor points; and wherein, the curve fitting module is further configured to determine a second set of parameters associated with the learning curve based, at least in part, on the second confidence score and the weight vector. ### Claim 22 (depends on 21) The prediction model server of claim 21, wherein the first set of parameters include average length of a segment, average length of tokens, lexical diversity, word-order divergence. ### Claim 23 (depends on 21) The prediction model server of claim 21 further comprising: a benchmark module configured to generate a first curve to determine a first score for each of the first set of anchor points; a regression module configured to generate a regression model, wherein the weight vector is assigned to each of the first set of anchor points using the regression model based on the first set of parameters, and the first score; and the scoring module configured to compute a first confidence score based on the weight vector, the first set of parameters and the first score. ### Claim 24 (depends on 23) The prediction model server of claim 23, wherein the second set of parameters is determined based on the first confidence score. ### Claim 25 (depends on 21) The prediction model server of claim 21, wherein each of the first set of anchor points includes a different number of monolingual segments. ### Claim 26 (depends on 21) The prediction model server of claim 21, wherein each of the second set of anchor points includes a different number of bi-lingual segments.