📢 Organic listening of the voice of customer: free and online tool.

New: API access now available

Try it, just click on the link:

https://nocodefunctions.com /api/organicForAText? text-lang= en & text= Buy our products now & explanation= on & output-format= html & explanation-lang= fr

You can modify these parameters in the url:

text-lang: the language of the text to anlayze. "en" for English, "fr" for French or "es" for Spanish
text: the text to analyze
explanation: this parameter has two values: "on" for rich explanations, or "off" if you just need the result without explanations
output-format: choose between 3 values: "plain", "html" or "json"
explanation-lang: choose the language of the explanations. The parameter is a two-letter code for the language ("zh" for Chinese, for instance)

... or read below about what it does, and how it works:

Analysis of organic voice of customer for Twitter, Instagram and beyond

This function classifies your texts in two categories:
organic: when the text is written in a style which is natural, personal, individual.
promoted: when the text is written in a style which is corporate, sponsored, artificial.
It works best on social media such as tweets for Twitter, comments on Instagram posts and other very short texts in English and French.
Born in 2020, this tool is under continuous development.

🚀 Fast and performant

The tool is programmed with Java which performances are similar too, or better than Python. It does not rely on part-of-speech tagging (POS tagging), which makes it even faster. Parallel computing is used in subfunctions and great care has been given to performance.

The model - short description

The function examines each term of the text and applies a series of rules (in this context, is the term organic or promoted?). It also considers emojis, punctuation, hashtags, capitalized words... to determine the result.

The model - long description

The principles followed by the tool are similar to the function for sentiment analysis, except for the rules to be applied. The tool follows these steps:

check on the length of the text: if it is just a couple of words short, it will not be classified
removal of urls, removal of content in quotes, normalization of apostrophs
two versions of the text are established: one where all the accents and special characters are removed, and one where they are retained. All the following steps will apply to both versions of the text.
check on emojis, emoticons and onomatopeia. Do not preprocess your text to remove them, they provide useful information on sentiment!
check on hashtags, if any
decomposition of the text in n-grams up to four-grams
for each n-gram:
- skip it if it belongs to a pre-established list of stop words
- check for repeated characters and remove them as necessary (yeeeahhh! becomes yeah!)
- check if the n-gram is contained in the pre-established list of vocabulary / expressions / style signalling "corporate speech". If so, the corresponding rule is applied.
final decision, based on the results of the previous steps.

Import your data, get the results: an example

Organic vs promoted content: further considerations

This tool identifies markers of corporate speak. This characterization is not clearly defined in the academic literature. The intent is to provide content managers and analysts a tool which allows them to filter out content which does not represent a genuine voice of the customer, such as:

calls to action: "VOTE", "BUY", "PLAY", ...
promotional invitations: "hottest deals", "give away", ...
expressions of opinion which are crafted by a communication specialist, as hinted by the specific terms used: "unveiled", "stoked to", "reinvents" ...

This is work in progress and we welcome your feedback and suggestions (analysis@exploreyourdata.com). We believe this function is already extremely useful, helping analysts specializing in opinion mining clean their dataset from the content which does not originate from individuals / customers but from corporations and communication officers (check our function for sentiment analysis). This will make their measurement more reliable, especially for opinion mining.

or

or