language-learning
language-learning copied to clipboard
singnet
OpenCog Unsupervised Language Learning
The goals of Unsupervised Segmentation Learning (USL) are: 1) Unsupervised learn of lexicon and tokenisation for languages like Chinese 2) Unsupervised learn for sentence splitting for languages like Chinese 3)...
_This MSc student research task may or may not end-up with actual code submitted into this repository_ Main Goal: **Exploration of applicability of explainable artificial intelligence technique for sentiment analysis...
After PR https://github.com/singnet/language-learning/pull/258, the "mini-pipeline" ppln.py that does grammar-learning and parse-evaluation stopped working in my system. I get the following error: ``` Grammar Learner + Tester ppln v.0.0.1 started 2019-10-08...
Need to add costs to rules based on statistics for better parsing with account to these costs. First, this has to be done for GL ILE algorithm and we see...
Currently, the main parse-evaluator doesn't ignore reference sentences with no links when calculating the overall performance metrics (precision, recall, F1). Given this situation, if a reference parse comes with no...
**Specification** This task address complication of the Unsupervised Language Learning (ULL) problem (specifically, Unsupervised Text Parsing) with the following conditions: 1. The input (training) streams of symbols is not segmented...
Two problems: 1) Inconsistent rounding: http://langlearn.singularitynet.io/data/aglushchenko_parses/GCB-NQ-dILEd-MWC-MSL-2019-04-24/GCB-NQ-dILEd-MWC-MSL-summary.txt 0 2 99.60% 1.00 Need to round F1 to 4 decimal places after period so the rounding appears consistent. 2) Some words in the...
As discussed with @linas in email thread and @OlegBaskov and @glicerico in Slack: Currently, Grammar Learner (GL) clusters words into word categories or Link Grammar (LG) rules but it does...
"Content-word" parse-evaluator mode was implemented in "alternative" evaluator in PR: https://github.com/singnet/language-learning/pull/237. It needs to be included in main parse-evaluator at some point, also.
Need to see how to make MST-Parses on large corpora to work faster, below is the discussion: @glicerico : It now takes around 100 hours to perform MST-parses using the...