THE API

In addition to the lexikat.com platform, we also offer our text analysis systems in API format. Because our dictionary is designed for capturing conceptual links as humans perceive them, it achieves far higher recall and precision than other models.

Algorithm          

 

Lexikat (upper and lower case)

NLTK (uncased)

Gensim Word2Vec (upper and lower case, trained on Google news corpus)

GloVe model (uncased, pretrained using Wikipedia 2014 and Gigaword 5)

Words related to “apple”

 

almond, pear, apricot, melon, apples, onion, strawberry, artichoke, orange_peel, berry, cucumber, eggplant, umeboshi, omelette, raspberry, ice_cream_cone, peach, watermelon, lemon, apple_tree, avocado, cherry, omelet, egg, turnip, pie, unripe, edible_seaweed, asparagus, acorn, pomegranate, pumpkin, edible_fruit, honey, tomato, potato, strawberries, carrot, herb, spinach, aniseed, thyme, cabbages, fruit, orange_tree, celery, cabbage, agave, radish, chicken, Microsoft, Netscape, Apple_Inc., Intel, BlackBerry, Apple_Computer, Google, Sony, Adobe, Asus, iPhone, Red_Hat, Atari, HTC, NeXT, Novell, Nvidia, Chrome_OS, Xoom, iPhone_5S, Motorola, iPhone_5, AMD, iPod, Iomega, Autodesk, iPad, Macromedia, Symantec, Palm_Pre, Verizon_Wireless, iMac, iPhone_4, NVIDIA, PCjr, Macintosh, Windows_8, Symbian, SanDisk, Windows_Phone, TouchPad, webOS, WebOS, Galaxy_Nexus, VMware, Corel, Apple_iPhone, Lenovo, PageMaker, Mozilla

apple, apple, orchard_apple_tree, malus_pumila

apples, pear, fruit, berry, pears, strawberry, peach, potato, grape, blueberry, Apple_AAPL, Apple_Nasdaq_AAPL, Apple_NASDAQ_AAPL, Apple_Computer, iPhone, Apple_NSDQ_AAPL, Steve_Jobs, iPad, Apple_nasdaq_AAPL, AAPL_PriceWatch_Alert

iphone, macintosh, ipod, microsoft, ipad, intel, ibm, google, imac, software

We also offer every component of our software as an independent API for use in your coding and data analysis projects.

Our topic modelling relies on a multi-million word crowdsourced database to provide far more human-like results than any alternative currently available. For every word in our dictionary we search the web to identify words and concepts that appear more frequently on the web pages that mention that word most frequently.

 

Who made our database? You did. Every time you run a web search or contribute to a website, you improve our results.  

We can provide:

- English language tokeniser

- English language tokeniser with multi-word and common phrase identification

- English language data cleaner

- English language topic modeling

- English language topic dictionary

- English language sentiment analysis

- Chinese language tokeniser

- Chinese language tokeniser with multi-word and common phrase identification

- English language data cleaner

- Chinese language topic modeling

- Chinese language topic dictionary

- Chinese language sentiment analysis

For more information, or to request a free trial, please contact sales@lexikat.com.