HistCite software import documents

Transferred from: Fan Spring and Autumn

The previous blog introduced some situations of HistCite software and solutions to a problem in use. This article describes how to use this software. In fact, the most important thing is to import the literature. The rest is to view various sorting and generate reference relationship graphs to lock important documents.

   The document imported by HistCite must come from the text file of the WoK database. The method for exporting the document from the WoK database is as follows:

   First, go to the Web of Science database. Do not start searching as soon as you enter WoK. That is, all databases are searched, and the exported documents are duplicated, which will increase the export workload, to enter the Web of Science (third tab), as shown in figure

   For example, for system identification, system identification. The result is as follows. We can see that 156560 results are found. This result is too many and the export workload is huge. Therefore, it is necessary to refine it, you can refine the search results based on the refining entries on the left. In addition, you can click the analysis search result (the link on the right) to perform a simple analysis on the literature, and then determine the refining method.

   After refining, you can export the data. The following export options are available at the bottom of the page:

   You need to make the following settings (a maximum of 500 records can be exported at a time). in step 1, select the full record and reference documents, select the Save format as plain text, and click Save.

   The following processing box is displayed. After processing 500 records, the "save" dialog box is displayed automatically.

   Then, you can return to the previous page to continue exporting the subsequent documents.

   When using the browser, it is found that 2nd export times will crash. If you use a google browser, this will not happen.

 

   The above documents are exported from the WoS v5 database and cannot be directly imported to HistCite. The following processing is required (the v4 version is said to be acceptable, but the v4 version searches for the same keywords, fewer search results than v5 ):

   Open the saved plain text file. Note that you need to use the advanced tool to open it. The laptop that comes with windows cannot complete the conversion process. For example, use Notepad ++ to open it. Click ctrl + f to switch to replace, then, replace ER \ n\ nnn. note that you must select the full-word match and extension options for the settings. Otherwise, the \ n (that is, the carriage return character) will not be recognized, click replace all. Of course, you can open all exported documents and click "replace all open files" to complete the replacement once.

   Then you can import the data to HistCite,

   Open HistCite. A black box is displayed as follows. The software will build a server on the local machine and open http: // 127.0.0.1: 1925/in the default browser /, this is the work URL of HistCite. You can enter this URL in other browsers.

   We recommend that you use the IE or google browser to open the web page, as shown below:

   Then you can import the document, but you need to set it. If it is IE8 or higher, you need to open the Custom Level in the Security tab of Internet Options, enable the "include local directory path when uploading files to the server" option, as shown below:

   Or add the trusted site http: // 127.0.0.1, as shown below:

   If it is a google browser, you also need to switch to the compatible mode, as shown below:

   There are two ways to import the document: 1: Select Add File ..., Add the text document you just downloaded

   2: drag the text document to the HistCite icon and import it directly. This method is suitable for importing multiple files together.

   Note that the documents exported by WoS v5 sometimes have problems, as shown below:

   The cause of this problem is unknown, and the WoS will probably improve the bug later. Now we can barely solve this problem by deleting about 60% of the documents in the document text file, the text format is roughly as follows. The first two rows are identifiers. PT to ER identifies a document record. EF indicates the end of the document.

FN Thomson Reuters Web of Knowledge
Virtual Reality 1.0
PT J

...

ER

 

PT J

...

ER
EF

  Then the import is successful, as shown below:

   This problem is estimated to be caused by a problem in the format of a document in the file, so that the document can be imported and deleted. However, it is difficult to locate the document accurately, just try to delete dozens of articles!

   The test shows that if you import 6 or 90% documents, there are only a few files with problems. The above method does not solve the problem of importing documents.

  

   Then we can perform citation analysis normally. We can sort data by Graph Maker under Tools... You can visualize the citation relationship as follows:

   If you have imported a lot of document text, you can save it as an hci file for centralized storage. Next time you don't have to re-import that much text.