"User Research" example based on tourist review value mining examples

Editor: Reading customers is extremely important for user research. The author of this article shared the value of information based on tourist reviews to do the relevant content of user research, telling the process and steps of tourist research, etc. Essence

Following the previous article, I share how to understand the user from the theoretical and tool recommendation. This article will be fully presented from the beginning to the end to the appearance of the user's comment to the content analysis, in order to have the analysis of the review analysis Students with demand provide more detailed help. If you do n’t read the previous article, you can click on my personal homepage to view.

In order to make the text more convenient, this article will share the tourism industry with rich comments and loose anti -climbing measures. If I am the manager of Feng Xiaogang's Film Commune, in the face of a large number of tourist comments on various OTA platforms, I started my own tourist research journey.

1. Getting of the content of the review content

According to browsing, the online reviews of the scenic area are concentrated on several OTA websites, namely Ctrip, Qunar, Tuniu, and Meituan.

The first step is to collect the scenic area on the website of this platform, and collect the content of the review content.

Ctrip.com as an example.

The second step is to open the Python, write the code (there are many code on the Internet that can be directly called), configure the number of pages you need to crawl and the fields you expect to crawl, and start collecting. If you need to comment on the source code of Ctrip, leave a message in the comment area.

If it is a non -programming method, the Houyi collector here is an example. After installation, enter the URL, click smart collection, wait for the loading page, and edit the bottom of the page to remove the unwanted fields. The user ID, comment content and reviews are retained in the article. time.

After the crawling is completed, generate the excel form. When you open the form, you can see that the collection has been collected and a total of 3900 pieces are obtained. At this point, the data collection phase is over.

Step 2. Pre -processing data

Words and removal words. Open the Rost CM6 software and find that it only supports the content of the text format, so the Excel table is directed as the text format. Then open the word split window in the interface of the Rost CM6, select the file to be treated, select the result to save the position, select the word table, and then determine the operation.

Third, the third step, frequency analysis and emotion analysis

The same is the rast cm6. Open the frequency analysis window, select the segmentation result output in the previous step, determine the word division, and obtain the excel file sorted by the frequency of words. The following table.

From the contents of the table, the style of Feng Xiaogang's film commune in the minds of tourists is at the level of the positioning of the scenic spot, that is, "Fanghua", "Nanyang", "Republic of China", "Crossing" in high -frequency words, etc. Corresponding to words, you can also see that "Hainan" and "Haikou" representing local characteristics are repeatedly mentioned.

Secondly, most tourists' perception of the entertainment experience of the scenic area is concentrated on "taking pictures". Interactive environment shaping, at the same time, from "tickets" and "fares", you can see that tourists value the sales of tickets in the scenic area. In addition, "night" and "night views" frequently appear, indicating that compared with the daytime scenic area, Feng Xiaogang in the evening at night, Feng Xiaogang in the evening The film commune has another characteristic.

Therefore, it is cared for by tourists, and the "Ice and Snow" world in the scenic area and the "Aquarium" also leave a lot of impression on tourists. In addition, in the emotional sense of high -frequency words, I see "worth", "fun", "" fun "," The words such as good -looking "," convenient ", and other opposite perceptions such as" not worth "," charging "," too expensive "," cheap "can be seen. Has a completely different effect.

At this step, the most basic text analysis can be achieved, and then the word frequency analysis is achieved, and then in order to make the words more visually visual, it is imported into the word cloud website recommended by the previous article to generate the word cloud.

Still in the Rost CM6 domain, open the emotional analysis window, introduce the comment text of the unlimited word, and one -click analysis. After simple processing, the tourists from Feng Xiaogang Film Commune are commented as follows. It can be found that tourists' positive emotional ratio in Feng Xiaogang's film commune is greater than negative emotions. The overall emotional bias is roughly positive, and the extremes of emotions are not significant.

Fourth, the fourth step, the semantic network together

For convenience, this article only generates the most basic semantic network diagram in the ROST CM6. The generating process does not go into details. It is similar to the frequency analysis and emotional analysis mentioned above. The semantic network diagram is as follows (this article produces the semantic network diagram according to positive and negative emotions).

1. Positive emotions

The positive emotions expressed by Feng Xiaogang Film Commune are mostly good -looking, fun, suitable, etc. These positive emotions are combined with the service positioning of the scenic destination.

It can be seen from the semantic network composition of positive emotions that "taking pictures" as one of the central nodes, closely linked to words such as "cheongsam", "clothes", "clothing", etc. At the same time The entertainment activity is mixed with many of the clothing in the scenic area in the experience of tourists.

Then see the word groups with "architecture" as the center, vocabulary such as "Nanyang", "Republic of China", "Crossing", and "Times", which is completely consistent with the architectural style of the scenic area; "The vocabulary of the central words is mostly associated with" Commune "," Feng Xiaogang "," shooting ", and" small courtyard ". It can be considered that in the scenic area, the status of Fanghua Courtyard is not light. Essence 2. Negative emotions

Visitors show more annoyance and regret in negative emotions. The cause of it is related to the price management of the ticket price and the content management of the scenic spot and the tourist's own cognition.

The unreasonable setting of ticket prices in scenic spots has led to poor experience in the experience of tourists, resulting in negative emotions. From the semantic network structure diagram of negative emotions (right), it can be seen that the main negative emotions are shown in the words "ticket" and "scenic spot" as the node.

For example, the word "still" and "inconvenience" connected to "tickets" reflect the serious phenomenon of repeated charges in the park, and stimulates the dissatisfaction of tourists. "Tickets" and other words effectively reflect the area and content of Feng Xiaogang Film Commune for some tourists who produce negative emotions, which makes tourists have a regrets that the scenic spot does not return the fare without returning the fare.

Fifth step, fifth, theme analysis

This article uses the theme word clustering analysis based on the TF-IDF method. TF means word frequency, which is used to calculate the frequency of the vocabulary entries. IDF means anti-text frequency to measure the general importance of the aforementioned entry.

Its calculation formula is:

Tf -df = TF × IDF

The online review text that has been investigated above has been processed into the XLSX format to import python, and the LDA theme model is built. The TF-IDF method is used. Repeated test adjustment vocabulary to the highest threshold is 0.4 (that is, the word is more than 40%of more than 40% Over the comment, it is abolished, thinking that it has no characteristic significance), and the number of clustering themes is 4. The final scrutiny of the online review topic is as shown in the following table.

The results of the theme words show that tourists' perception of Feng Xiaogang's film commune includes four aspects: the main characteristics, entertainment facilities, infrastructure and emotional perception. It can be considered more facial and complex. And the theme cluster results are highly converged with the three categories of human brain's belonging, which is in line with expectations.

One of the attractions of Feng Xiaogang's Film Commune is the creation of the venue, which is manifested in the "Fanghua" and "Movies", "Commune" and "Courtyard", which are frequently showed in the tourist experience, proved that it has produced an important thing in the process of tourists tour. The influence and evaluation are mostly "good".

At the same time, you can see the "photography", "crossing", "the Republic of China", "Nanyang" and "Weather" that tourists experienced in the second subject. focus on.

In addition, from the emotional perception of tourists, the fourth theme can see the opposite emotional perception of "cost -effective", "worth", "like", "general", etc., can be considered to be considered to be different people. Different from the same or whether the price sensitive customers are different, the scenic area can set up ticket types from the perspective of price discrimination to take into account the feelings of different consumer groups.

Step 6, 6, conclusion

After the above -mentioned tourist reviews, the manager of Feng Xiaogang's Film Commune can summarize the following conclusions to support the improvement of the operation and management side.

First, improve the entrance service of the scenic area. The scenic spot should formulate a reasonable pricing strategy, such as considering the experience of tourists with different degrees of acceptance and realizing the differential pricing strategy. Within the acceptable range of scenic spots, different price standards such as ordinary tourists, students, children, and elderly people such as ordinary tourists, students, children, and elderly people are defined. Then there is the peak adjustment of the fare. The peak season of the tourism during the holidays may wish to be full price. In the off -season of tourism, the fare needs to be discounted. At the same time, the purchase of tickets for different channels should also be priced. At the same time, the price of the package is lowered to encourage tourists to buy discount packages to encourage tourists to experience the full picture of the scenic area. Second, tour guide service staff should be added in the scenic area to bless tourists for tourists, so that tourists who do not like to take pictures can also be immersed in the style of the scenic area and the story behind the building under the explanation of the staff. The emotion of prices has extended the time of playing such tourists. Third, strengthen market supervision and improve the quality of tourism services. Strengthen market supervision, strictly standardize the various market behaviors in Feng Xiaogang's film commune scenic spot, and formulate a unified standard for the service and quality of the product. Resolutely prohibit the emergence of slaughtering, especially to strengthen supervision of clothing rental services in the scenic area. To solve the problems encountered in tourists in a series of measures, reducing the negative emotions of tourists.

This article is published by @ 我 本 本 本 本 本 is the product manager. Reprinting

The title map is from Unsplash, based on the CC0 protocol

- END -