Hi I need a way to identify specific information. I have a dataset: company_name company_id brand_id brand_name word_count The brand_name contains either the brand itself or a catchphrase that promotes the brand.  ABC Corp  1234  A888888  Vallina Soda                                 2 ABC Corp  1234  A666666  Diet Soda Sugar Free                  4 ABC Corp  1234  A555555  Share a Soda Moment                 4 ABC Corp  1234  A333333  SodaRewards                              1 ABC Corp  1234  A222222  Soda                                            1 ABC Corp  1234  A000000  iSoda                                           1 ABC Corp  1234  A999999  Soda Original                               2 ABC Corp  1234  B888888  Drink                                            1 ABC Corp  1234  A888888  Drink Multi flavored                      3 I don't have a list of brands, so I need a way to identify brand keyword (e.g. Soda, Drink are the keywords) So for 1 word brand_name, two possibilities: brand itself (Soda) or word containing the brand (SodaRewards, iSoda).  2-word brand_names, two possibilities: 2-word brand (Best Quality, Higher Standards) or phrase containing 1-word brand (Vanilla Soda, Soda Original). 3-word brand_names, three possibilities: 3-word brand (Best Fried Chicken), phrase containing 1-word brand (Drink Multi Flavored), or phrase containing 2-word brand  (Best Quality Tools). 4-word brand_name, four possibilities: 4-word brand (The American's Finest Style), phrase containing 1-word brand, phrase containing 2-word brand, or phrase containing 3-word brand. and so on....The longest string contains n words. The search is done at the firm level. Also, most brands are short, the longer strings are usually catchphrase promoting the brand.  Catchphrase could contain the brand (obs.3) or not related to the brand.  Those catchphrases not containing the brand can be treated as a separate brand. I need a way to search for brand keywords using the algorithm, so that the output will be something like company_name  company_id  brand_id    brand_name                       keyword ABC Corp           1234              A888888   Vallina Soda                       Soda ABC Corp           1234              A666666    Diet Soda Sugar Free       Soda ABC Corp           1234              A555555    Share a Soda Moment      Soda ABC Corp           1234              A333333    SodaRewards                   Soda ABC Corp           1234              A222222    Soda                                 Soda ABC Corp           1234              A000000    iSoda                                 Soda ABC Corp           1234              A999999    Soda Original                    Soda ABC Corp           1234              B888888    Drink                                 Drink ABC Corp           1234              A888888    Drink Multi flavored           Drink Can some one guide how can i achieve this. ... View more