Part 3.
3. A couple of words about semantic differential and connotative meaning of textsNow the fun begins. The attitude towards security services is rather ambiguous all over the world. Some consider them defenders of fundamental human values and rights, fighters against the world’s evil. Others, on the contrary, consider security services the main violators of fundamental rights, trying to monitor everything they can in order to put all the processes in society under control. We will try to use a part of the tools used by the special services for “peaceful purposes”.
A well-known American system Prism, used by the USA security services, is designed among other things for automatized monitoring of public mood in the Internet. 24 hours a day, Prism collects user comments on a large number of indicative (that is, for whatever reasons, currently relevant) topics from all open sources. And analyzes the attitude of users to these indicative topics. Interested people can study in more detail everything connected with the
semantic differential and
connotative meanings of texts. We will briefly explain the essence.
Each word, used in writing or orally, except for the usual and well-known meaning, also has the so-called connotative meaning. A whole set of connotative meanings, to be more precise. Connotative meanings can characterize a person’s real attitude to a topic, show how much person is interested, and reveal how much a topic induces to actions. In total, more than 20 different parameters describing the attitude of a person to the commented topic can be analyzed. User could comment on topic «The USA Democratic Party» something like that: «The Democratic Party – is the best party in this country! Democrats are our future. I wish to join democrats right now. Nothing can be better that Democratic Party. Democratic Party forever! ». Technically the text suggests that the user is keen on democrats. But analysis made with regard to exactly which words and in what order were used to express these thoughts can show that a person is not crazy about democrats at all. And even ready to do some harm to the democrats on occasion. Thus, the analysis of connotative meanings of words using the semantic differential technique allows us to reveal the true attitude of a person to the subject.
With the help of such a comprehensive analysis security services monitor, among other things, revolutionary sentiments in society. At the moment we are not interested in monitoring the public image of power in the countries of the Middle East and the degree of revolutionism in the countries of South America. But the
implementation of semantic differential for the analysis of investors’ real attitude to current ICOs sounds intriguing. The described method fits perfectly for our purpose.
The above-mentioned ICO Catalog collects users’ comments on current ICOs 24 hours a day and then analyzes each ICO by three parameters:
• Evaluation – users’ attitude to ICO on a scale from -3 (the most negative attitude) to +3 (the most positive attitude);
• Activity – shows willingness of users to take any actions related to this ICO (to buy token, to leave a comment on forum, etc.) on a scale from -3 to +3;
• Potency – shows how much users are interested in ICO on a scale from -3 (minimum interest) to +3 (maximum interest).
Having processed a huge array of statistical data, containing Evaluation, Activity and Potency parameters, collectively forming a group of so-called EAP-parameters, we obtained interesting data, which turned out to be nothing like “white noise”.
a. Potency ParameterLet us start from Potency, the last parameter.
Pic. 6. The dependence of amount of funds raised during ICO (as percentage of the declared ICO Hard Cap) on Potency parameter.Diagram in pic. 6 illustrates demonstrates dependence between declared Hard Cap percentage of funds raised during and Potency parameter. On the horizontal axis the data is divided according to groups of ICOs:
• Potency parameter is within the range of -0,06 to -0,05;
• Potency parameter is within the range of -0,05 to -0,04;
• Potency parameter is within the range of -0,04 to -0,03;
• And so on to the group of ICOs, with Potency parameter within the range of 0,11 to 0,12.
The vertical axis shows the average percentage of collected funds for each group of ICOs from the number of messages. The size of yellow circles on the diagram illustrates average amount of funds in millions of USD raised by ICOs from certain group. The trend line is marked in blue.
As expected, ICOs with Potency less than -0,02 did not raise even 10% of their Hard Caps. The total amounts of collected funds in US dollars for such ICOs are also characterized by small values. The success level of ICOs with Potency parameter exceeding -0,01 can be widely interpreted. However, the pattern of consistently high ICO success for high values of the Potency parameter is clearly visible. The total amounts of collected funds for ICOs with high Potency also have tendency to increase along with the growth of Potency parameter.
A short intermediate conclusion for
ICO organizers based on the analysis of the diagram at pic.6 – organizers can’t directly influence the potency parameter, as it reflects the true interest in the ICO. However, reduction of Potency value to the level less than -0,01 during ICO is a serious indicator for making changes in the marketing campaign. It is a sign of lack of interest in ICO project and close a call that ICO may not raise even 10% of the announced Hard Cap.
A short intermediate conclusion for
ICO investors based on the analysis of the diagram at pic.6 – study the Potency graph of ICOs at
ICO Catalog and avoid investing in projects with average Potency parameter remaining less than -0,01 for a long time.
b. Activity ParameterCertainly, Potency parameter characterizing the real interest of the audience in ICO is the most important among other EAP-parameters. The Activity parameter, belonging to the same group and characterizing the willingness of users to take actions related to the analyzed ICO, gives a vaguer picture. First of all, it’s because activity does not always reflect willingness to buy tokens. Desire to show some ICO-related activity may also manifest in the intention to leave a post in the ICO topic (not always positive, by the way), or closing the page with ICO description – the variability of the possible activity is quite high.
Pic. 7. The dependence of amount of funds raised during ICO (as percentage of the declared ICO Hard Cap) on Activity parameter.Diagram in pic. 7 illustrates dependence between declared Hard Cap percentage of funds raised during ICO and Activity parameter. On the horizontal axis the data is divided according to groups of ICOs:
• Activity parameter is within the range of -0,07 to -0,06;
• Activity parameter is within the range of -0,06 to -0,05;
• Activity parameter is within the range of -0,05 to -0,02;
• And so on to the group of ICOs, with Activity parameter within the range of 0,17 to 0,18.
The vertical axis shows the average percentage of collected funds for each group of ICOs from the number of messages. The size of yellow circles on the diagram illustrates average amount of funds in millions of USD raised by ICOs from certain group. The trend line is marked in blue.
Probably, for a more detailed analysis of the diagram shown in pic. 7, it is necessary to use additional filters and conditions imposed on the initial data used for the analysis. Another option is using the cumulative parameters, constructed based on combinations: [Activity, Potency], [Activity, Evaluation] or [Activity, Potency, Evaluation].
For
ICO organizers we can give cautious recommendation to keep track of Activity parameter during ICO and if it drops below -0,08, carefully make some corrections in the marketing campaign. Of course, from the graph in pic. 7 it does not follow that a low value of the Activity is expressly bad. However, the lack of desire even to write in the ICO topic is also difficult to call a good thing.
Short hint for
ICO investors: as for the Activity parameter, it should also be recognized that it, apparently, is also a secondary parameter and should be considered in conjunction with other ICO parameters.
c. Evaluation ParameterStrangely enough, the Evaluation parameter, which characterizes the real attitude of users to the ongoing ICO, gives the most unexpected picture. The diagram in pic. 8 shows a clear shift of a part of the data to the left side with respect to 0 of the horizontal X axis.
Pic. 8. The dependence of amount of funds raised during ICO (as percentage of the declared ICO Hard Cap) on Evaluation parameter.Diagram in pic. 8 illustrates dependence between declared Hard Cap percentage of funds raised during ICO and Evaluation parameter. On the horizontal axis the data is divided according to groups of ICOs:
• Evaluation parameter is within the range of -0,10 to -0,09;
• Evaluation parameter is within the range of -0,09 to -0,08;
• Evaluation parameter is within the range of -0,08 to -0,07;
• And so on to the group of ICOs, with Evaluation parameter within the range of 0,04 дo 0,05.
The vertical axis shows the average percentage of collected funds for each group of ICOs from the number of messages. The size of yellow circles on the diagram illustrates average amount of funds in millions of USD raised by ICOs from certain group. The trend line is marked in blue.
The existing shift of some of the values along the X axis to the left of 0 can be caused by various factors, such as:
• The overall shift of the estimates caused by the shifted scale in the Evaluation calculating algorithm itself;
• The presence of a large number of low-quality comments, ordered by the ICO organizers themselves, in the messages analyzed for the calculation of the Evaluation parameter. They ordered, of course, positive comments to liven the topic. The literal meaning of the comments ordered was positive and the ICO organizers accepted the work. The system simply shows the true attitude to the ICO from people who carry out the order for the topic. This option seems the most plausible;
• Poor attitude to something does not always mean lack of desire to make money on this “something”;
• The combination of the above-mentioned factors.
For now, we refrain from recommendations on use of the Evaluation parameter.